mirror of
https://github.com/fabro-sh/fabro.git
synced 2026-10-06 02:48:25 +00:00
114 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Bryan Helmkamp
|
9b5502bf9b
|
Fix edge thread_id precedence to match fidelity behavior
Edge thread_id should override node thread_id, consistent with how resolve_fidelity already works. The previous order was reversed. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
ba9988d4ee
|
Destructure ModelInfo to avoid unnecessary clone
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
c6f979e2f2
|
Resolve model aliases to canonical IDs in workflow graph nodes
ProviderInferenceTransform only inferred the provider but passed the raw alias (e.g. "gpt-54") to the LLM API, causing request failures. Rename to ModelResolutionTransform and resolve aliases via the model catalog so the canonical ID (e.g. "gpt-5.4") is used in API calls. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
003359b079
|
Unify goal prefix stripping into shared strip_goal_decoration() in fabro-util
Three places stripped markdown headings and `Plan:` prefixes from goals with slightly different logic. Extract a shared function so all call sites behave consistently, and fix `fabro run` which wasn't stripping at all. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
brynary-fabro[bot]
|
956e681e4f
|
Unified WorktreeSandbox (#117)
This PR introduces a unified `WorktreeSandbox` type in `fabro-sandbox`
that consolidates previously duplicated git worktree management logic
spread across `parallel.rs` and `run.rs`. The new type wraps any
`Arc<dyn Sandbox>`, handles the full worktree lifecycle (branch
creation, `worktree add`, and cleanup) in its `initialize()`/`cleanup()`
methods, overrides `working_directory()` and `exec_command()` to default
to the worktree path, and delegates all other `Sandbox` methods to the
inner sandbox. A `WorktreeConfig` struct controls behavior (branch name,
base SHA, worktree path, and a `skip_branch_creation` flag for resume
flows), and a `WorktreeEventCallback` mechanism bridges lifecycle events
to the workflow event system via a new
`EventEmitter::worktree_callback()` helper.
The old private `WorktreeSandbox` struct in `parallel.rs` (which only
redirected `exec_command` working dirs with no lifecycle awareness) is
removed and replaced with the shared implementation. The
`setup_worktree()` function in `run.rs` is also removed; its logic is
absorbed directly into the `SandboxProvider::Local` branch of sandbox
construction, where `WorktreeSandbox::initialize()` is called and
`std::env::set_current_dir()` follows on success. The resume path
(`run_from_branch`) similarly replaces direct `git::replace_worktree`
calls with `WorktreeSandbox` using `skip_branch_creation: true`. The
`MockSandbox` in `test_support.rs` gains `captured_commands` and
`captured_working_dirs` vectors to support sequenced-command assertions
in the new unit tests.
The `MockSandbox` enhancement is a notable improvement for testability
beyond this specific change—having the full ordered sequence of commands
rather than just the last one makes it straightforward to assert on
multi-step git workflows. One subtle behavior worth noting is that in
`parallel.rs` the `git reset --hard` step previously present after
worktree creation is now absent from `WorktreeSandbox::initialize()`;
the plan mentioned it but the implementation deliberately omits it (the
branch is already force-set to the target SHA, so the reset was
redundant for the parallel case). Cleanup for parallel branches
continues to go through `engine::git_remove_worktree` on the parent
sandbox rather than calling `wt_sandbox.cleanup()`, since the sandbox
`Arc` is consumed by the spawned task—this is a reasonable tradeoff
noted in the plan.
### Fabro Details
<details>
<summary>Ran 11 stages in 62m 32s for $3.67</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 1m 11s | – | 0 |
| preflight_lint | 12s | – | 0 |
| implement | 30m 15s | $2.15 | 0 |
| simplify_opus | 15m 5s | $0.71 | 0 |
| simplify_gpt | 11m 8s | $0.54 | 0 |
| verify | 46s | – | 0 |
| fixup | 3m 17s | $0.27 | 0 |
| verify | 46s | – | 0 |
| fmt | 1s | – | 0 |
| **Total** | **62m 32s** | **$3.67** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
3680a57a25
|
Decompose fabro run into create / start / attach (#116)
This PR decomposes `fabro run` into three composable primitives —
`create`, `start`, and `attach` — following the Docker-style lifecycle
model. Previously, `fabro run` performed everything in a single
monolithic function, and `--detach` was implemented by reconstructing
CLI argv to spawn a child process, which was brittle and hard to extend.
The new architecture cleanly separates concerns: `fabro create`
allocates the run directory and persists a `RunSpec` struct to
`spec.json`; `fabro start` spawns a detached `_run_engine` process (a
hidden internal command that reads `spec.json`) via `setsid`; and `fabro
attach` tails `progress.jsonl` with live rendering and handles
file-based interview IPC. `fabro run` is now a composition of these
three primitives, and `fabro run --detach` simply skips the attach step.
The main rendering work lives in a new `handle_json_line()` method on
`ProgressUI` that parses JSONL envelopes and dispatches to the same
internal rendering methods already used by the in-process event handler.
This preserves 100% rendering fidelity without duplicating
spinner/stage/tool-call logic — the attach loop just feeds file lines
into the same code paths. File-based interview IPC is handled in the
attach loop itself: it watches for `interview_request.json`, prompts the
user via `ConsoleInterviewer`, and writes `interview_response.json` back
for the engine to consume. The `hide_bars`/`show_bars` methods
previously private to `ProgressAwareInterviewer` are promoted to public
methods on `ProgressUI` and reused in both the attach loop and the
existing in-process interviewer.
The old `detach_run()` function in `main.rs`, which reconstructed argv
by string-scanning `std::env::args()`, is deleted entirely and replaced
by the `create` + `start` composition. New tests cover the
`handle_json_line` dispatch paths (stage started/completed, tool calls,
retro events, invalid input) and the CLI argument parsing for the new
command variants.
### Fabro Details
<details>
<summary>Ran 9 stages in 30m 55s for $8.55</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 1m 11s | – | 0 |
| preflight_lint | 12s | – | 0 |
| implement | 18m 15s | $5.28 | 0 |
| simplify_opus | 10m 31s | $3.27 | 0 |
| simplify_gpt | 0s | – | 0 |
| verify | 17s | – | 0 |
| fmt | 1s | – | 0 |
| **Total** | **30m 55s** | **$8.55** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
863ecf7c22
|
Emit StageStarted on retry attempts (#113)
This change fixes a bug where the CLI progress UI would freeze during
stage retry attempts. When a stage fails with a transient error and the
engine retries it, the UI was never notified that a new attempt had
begun — `StageStarted` was only emitted once before the retry loop, so
subsequent attempts had no corresponding entry in `active_stages` and
all their progress events were silently dropped.
The fix moves `StageStarted` emission inside the retry loop for attempts
after the first. The first attempt's emission stays in its original
location (before the `StageStart` lifecycle hook) so that skipped nodes
still receive the event and hooks continue to fire only once. Each retry
now emits `StageStarted` with the correct `attempt` and `max_attempts`
values, which the existing `on_stage_started` handler in the progress UI
already handles correctly by inserting a fresh `ActiveStage` entry and
creating a new spinner.
A regression test is included that wires up a
`FailOnceThenSucceedHandler` — a handler that returns a retryable error
on its first call and succeeds on the second — and asserts that exactly
two `StageStarted` events are emitted for the retried node, one per
attempt. This directly encodes the invariant that every attempt,
including retries, produces a visible `StageStarted` event.
### Fabro Details
<details>
<summary>Ran 9 stages in 14m 4s for $2.78</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 1m 0s | – | 0 |
| preflight_lint | 10s | – | 0 |
| implement | 6m 38s | $1.54 | 0 |
| simplify_opus | 4m 38s | $1.23 | 0 |
| simplify_gpt | 0s | – | 0 |
| verify | 1m 10s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **14m 4s** | **$2.78** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
69ef57cd4a
|
Add node-level model validation + missing catalog aliases (#110)
This PR addresses two related issues that caused `fabro run` to fail at
runtime with an unhelpful LLM error when using model names like
`gpt-54`: the model catalog was missing hyphenated aliases, and the
validator had no rule to catch unknown model names on workflow nodes.
The catalog changes add `gpt-54`, `gpt-54-pro`, and `gpt-54-mini` as
recognized aliases alongside the existing `gpt54`-style variants for the
`gpt-5.4` model family. This makes the hyphenated form a valid input
that resolves correctly at both validation and runtime, and the updated
`fabro model list` output and insta snapshots reflect the expanded alias
columns. Three new alias-resolution unit tests confirm each mapping.
On the validation side, a new `NodeModelKnownRule` mirrors the existing
`StylesheetModelKnownRule` but iterates over graph nodes instead of
stylesheet declarations, emitting `Severity::Warning` diagnostics with
the relevant `node_id` when a node's `model` or `provider` attribute
doesn't match anything in the catalog. To avoid duplication, the inline
validation logic in `StylesheetModelKnownRule` was refactored into
shared `check_model_known` and `check_provider_known` helpers that both
rules now call. Five unit tests cover the valid-model, unknown-model,
alias, unknown-provider, and no-attribute cases. An unrelated dead-code
helper in `git.rs` was also removed as part of the cleanup.
### Fabro Details
<details>
<summary>Ran 11 stages in 12m 36s for $2.85</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 58s | – | 0 |
| preflight_lint | 10s | – | 0 |
| implement | 2m 56s | $1.13 | 0 |
| simplify_opus | 4m 12s | $1.13 | 0 |
| simplify_gpt | 0s | – | 0 |
| verify | 10s | – | 0 |
| fixup | 2m 27s | $0.59 | 0 |
| verify | 10s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **12m 36s** | **$2.85** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
62c310bcec
|
Extract fabro-model crate from fabro-llm (#109)
This PR extracts a new `fabro-model` crate from `fabro-llm`, isolating
the model catalog, provider identity types, and model metadata into a
focused, lightweight package. The new crate contains `Provider`,
`ModelId`, `ModelInfo` (and related structs), and the full catalog API
(`get_model_info`, `list_models`, `default_model_from_env`,
`build_fallback_chain`, etc.) — with only `serde` and `serde_json` as
runtime dependencies, no async runtime required.
The extraction follows a clean boundary: `fabro-llm` now depends on
`fabro-model` and re-exports `Provider` and `ModelId` at its root so
existing `fabro_llm::Provider` usage remains valid without a migration.
The `ProviderAdapter` trait, `StreamEventStream`, and
`validate_tool_choice` stay in `fabro-llm` since they depend on async
streaming types. All seven consumer crates (`fabro-agent`, `fabro-api`,
`fabro-cli`, `fabro-hooks`, `fabro-workflows`, `fabro-validate`, and
`fabro-llm` itself) are updated to import catalog functions and provider
types directly from `fabro_model` rather than `fabro_llm::catalog` or
`fabro_llm::provider`.
The most impactful dependency reduction is in `fabro-validate`, which
previously pulled in all of `fabro-llm` (tokio, reqwest, the full async
runtime) just to perform catalog lookups and validate provider names. It
now depends solely on `fabro-model`, eliminating that heavyweight
transitive closure. All other consumers retain their `fabro-llm`
dependency for client and streaming functionality, but gain a precise
import path for catalog and identity concerns.
### Fabro Details
<details>
<summary>Ran 9 stages in 21m 16s for $5.47</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 1m 10s | – | 0 |
| preflight_lint | 12s | – | 0 |
| implement | 13m 29s | $3.81 | 1 |
| simplify_opus | 5m 41s | $1.66 | 0 |
| simplify_gpt | 0s | – | 0 |
| verify | 15s | – | 0 |
| fmt | 1s | – | 0 |
| **Total** | **21m 16s** | **$5.47** | **1** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
Bryan Helmkamp
|
9b1cc75b20
|
Create fabro-sandbox crate, consolidating Sandbox trait and all implementations
Extract the Sandbox trait, types, and all sandbox implementations from fabro-agent and four separate crates (fabro-exe, fabro-ssh, fabro-sprites, fabro-daytona) into a single fabro-sandbox crate. This cleans up the dependency graph — implementation crates no longer pull in the full fabro-agent just for the trait. The new crate uses feature flags (local, docker, ssh, exe, sprites, daytona, test-support) to gate each implementation. The shell_quote() helper is unified into a single shared implementation, eliminating four duplicate copies. fabro-agent now re-exports all sandbox types from fabro-sandbox for backward compatibility. The four absorbed crates are removed. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
0f066b0a78
|
Fix agent sessions ignoring node-level model/provider from stylesheets
AgentApiBackend::create_session() always used the backend's default model/provider, ignoring attributes set on the node by stylesheet application. The one_shot path already read node.model() correctly but the agent session path (used by implement and other agent stages) did not. Also fixes usage reporting and provider_used.json to reflect the actual model used rather than the backend default. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
brynary-fabro[bot]
|
87403b19a5
|
Fix: Default provider should respect configured API keys (#103)
This PR fixes a bug where users with only OpenAI or Gemini API keys
configured would hit errors when running commands without an explicit
`--provider` flag, because the system hardcoded `Provider::Anthropic` as
the fallback in multiple places.
The core change introduces `Provider::default_from_env()`, which checks
which providers have API keys configured and selects the best available
one using a precedence order (Anthropic → OpenAI → Gemini), falling back
to Anthropic if none are found. A testable `default_with()` helper
enables thorough unit testing of the selection logic without relying on
environment variables. A companion `catalog::default_model_from_env()`
function pairs the provider selection with the correct default model
lookup.
All six call sites across the CLI, API server, workflow handler, and
agent that previously hardcoded `Provider::Anthropic` or used
`catalog::default_model()` without considering available keys have been
updated to use the new environment-aware defaults. The catalog functions
also now normalize provider aliases (e.g. `"open_ai"` → `"openai"`) to
prevent silent lookup misses.
### Fabro Details
<details>
<summary>Ran 10 stages in 28m 34s for $6.23</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 1m 8s | – | 0 |
| preflight_lint | 13s | – | 0 |
| implement | 4m 40s | $0.79 | 0 |
| simplify_opus | 7m 23s | $1.84 | 0 |
| simplify_gemini | 6m 24s | $1.44 | 0 |
| simplify_gpt | 7m 56s | $2.16 | 0 |
| verify | 14s | – | 0 |
| fmt | 1s | – | 0 |
| **Total** | **28m 34s** | **$6.23** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Fabro <fabro@fabro.sh>
|
||
|
Bryan Helmkamp
|
e658f9705e
|
Add skip_clone to DaytonaConfig to fix concurrent test failures
Two Daytona integration tests used std::env::set_current_dir to a temp directory so detect_repo_info() would fail and skip cloning. Since cwd is process-global, this poisoned concurrent tests. Replace with an explicit skip_clone config flag that skips repo detection and cloning during sandbox initialization. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
ce313f56a4
|
Remove dead ensure_clean_and_pushed function
This function had zero callers after sync_status was introduced in
|
||
|
Bryan Helmkamp
|
a5273e17a3
|
Add GitSyncStatus and WorkdirStrategy enums to fix cloud sandbox pre-run checks
The git sync check and auto-push logic was gated on should_create_worktree, which was always false for remote sandboxes. This meant Daytona/Exe/SSH runs silently proceeded without verifying commits were pushed or warning about uncommitted changes. Replace the git_clean boolean and should_create_worktree boolean with two enums (GitSyncStatus: Synced/Unsynced/Dirty and WorkdirStrategy: LocalDirectory/LocalWorktree/Cloud) so every combination is handled explicitly via match arms. Also display the base commit SHA for cloud runs. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
851edd2745
|
Fix private repo clone test to use current repo name
The assertion was checking for the old repo name brynary/arc instead of fabro-sh/fabro. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
36c0debbed
|
Fix asset collection tests by setting asset_globs
The Docker and Daytona asset collection tests were failing because asset_globs was empty, causing the engine to skip collection entirely. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
c57734244e
|
Move workflow CLI ownership into fabro-cli | ||
|
Bryan Helmkamp
|
165b817250
|
Fix fabro-config dependency inversion
Move config/data types from upstream crates (fabro-agent, fabro-mcp, fabro-workflows, fabro-hooks) down into fabro-config so it becomes a leaf crate depending only on fabro-util + external crates. New modules in fabro-config: - mcp.rs: McpServerConfig, McpTransport, McpServerEntry - sandbox.rs: DaytonaConfig, ExeConfig, SshConfig, SandboxConfig, etc. - hook.rs: HookEvent, HookDefinition, HookConfig, HookType, TlsMode - run.rs: RunDefaults, WorkflowRunConfig, LlmConfig, SetupConfig, etc. - project.rs: ProjectConfig, workflow discovery/resolution functions Source crates re-export from fabro-config for backward compatibility. Also removes stale strsim dep and moves toml to dev-deps in fabro-workflows. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
fab6603c40
|
Clean up fabro-hooks extraction
Extract set_hook_node() helper to deduplicate the 5 call sites that populate node fields on HookContext. The helper lives in fabro-workflows (which has the fabro-graphviz dependency) rather than fabro-hooks. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
b5e1d1fa74
|
Extract fabro-hooks crate from fabro-workflows
Move the self-contained hooks module (~2900 LOC) into its own crate to clarify the dependency graph and make the hook system independently reusable. The set_node convenience method is inlined at its two call sites in parallel.rs since it depends on fabro-graphviz types. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
51cfb55128
|
Clean up fabro-interview extraction
- Use already-imported names in run_from_branch instead of fully-qualified fabro_interview::* paths - Use std::io::Error::other() for serde error conversion (matches codebase convention, more concise) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
c18062bc3f
|
Extract fabro-interview crate from fabro-workflows
The interviewer module (trait + 7 implementations for human-in-the-loop interactions) had zero dependencies on fabro-workflows internals, making it a clean extraction. Consumers (fabro-api, fabro-slack) now depend on fabro-interview directly instead of reaching through fabro-workflows. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
ef56abb544
|
Clean up fabro-validate extraction
- Add From<ValidationError> for FabroError to eliminate duplicated .map_err(|e| FabroError::Validation(e.0)) at call sites - Use top-level `use` imports for stylesheet types in rules.rs instead of verbose fully-qualified paths - Remove duplicate parse_condition tests from fabro-workflows (already covered by fabro-graphviz) - Use //! inner doc comments in context/keys.rs Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
19e268424f
|
Extract fabro-validate crate from fabro-workflows
Move validation/lint framework and all 24 rules into a dedicated fabro-validate crate. As prerequisites, move Fidelity, stylesheet parser/types, and condition parser into fabro-graphviz (where the Graph types they operate on already live) so fabro-validate can depend on fabro-graphviz directly without a circular dependency. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
4c901f85d6
|
Extract fabro-daytona crate from fabro-workflows
Move DaytonaSandbox into its own crate, matching the pattern used by fabro-exe, fabro-sprites, and fabro-ssh. The new crate internalizes daytona_sdk::Client creation so callers never touch daytona-sdk directly: - new() is now async and creates the client internally - reconnect(name) replaces from_existing() + manual client/get boilerplate - daytona-sdk and daytona-api-client removed as fabro-workflows dependencies Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
693f52de62
|
Extract fabro-retro crate from fabro-workflows
Move retro.rs and retro_agent.rs into a new fabro-retro crate to reduce the size of fabro-workflows and clarify domain boundaries. Key design changes: - Add CompletedStage struct as a flat DTO that decouples retro derivation from Checkpoint/Outcome types in the workflow engine - derive_retro now takes Vec<CompletedStage> (owned) instead of &Checkpoint - run_retro_agent takes an event_callback closure instead of EventEmitter, pushing event filtering to the caller - Shared build_completed_stages() in fabro-workflows::lib converts Checkpoint → Vec<CompletedStage> for both run.rs and server.rs callers Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
49c7e329dd
|
Normalize fabro_graphviz imports to use shorter re-exported paths
Replace fabro_graphviz::graph::types:: with fabro_graphviz::graph:: everywhere, since graph/mod.rs re-exports types::*. Also simplify the From<GraphvizError> impl. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
b822a76f79
|
Extract fabro-graphviz crate from fabro-workflows
Move the self-contained graph/ and parser/ modules into a new fabro-graphviz crate so the Graphviz DOT parser can be used without pulling in the full workflow engine. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
brynary-fabro[bot]
|
0530605fdb
|
Add -q/--quiet to fabro ps (#64)
Adds a `-q`/`--quiet` flag to `fabro ps`, mirroring the behavior of
`docker ps -q`. When specified, the command outputs only full run IDs,
one per line, with no headers, footers, or "no runs found"
messages—empty output simply means no matching runs exist.
The quiet flag takes precedence over JSON output, and it composes
naturally with other flags like `-a` (e.g., `fabro ps -qa` lists all run
IDs).
### Fabro Details
<details>
<summary>Ran 10 stages in 10m 13s for $1.09</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $0.17 | 0 |
| simplify_opus | 0s | $0.30 | 0 |
| simplify_gemini | 0s | $0.30 | 0 |
| simplify_gpt | 0s | $0.32 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **10m 13s** | **$1.09** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
brynary-fabro[bot]
|
9d0a2a9a66
|
Move sandbox lifecycle into the engine (#22)
This PR moves sandbox lifecycle management (initialization, setup
commands, devcontainer phases, and cleanup) from the CLI's `run_command`
god function into the workflow engine. Two new engine methods are
introduced: `run_with_lifecycle()` orchestrates sandbox init, fires the
`SandboxReady` hook (now blocking by default), emits a new
`SandboxInitialized` event, handles remote git setup, runs setup
commands and devcontainer lifecycle phases, then delegates to the
existing `run_internal()` graph execution. `cleanup_sandbox()` fires the
`SandboxCleanup` hook and optionally tears down the sandbox. Both
`SandboxReady` and `SandboxCleanup` hook events were previously defined
but never fired — they now fire naturally within the engine alongside
all other hooks.
The CLI is simplified significantly: sandbox record persistence and
progress UI updates are handled via an event listener for
`SandboxInitialized` rather than inline code. The `run_from_branch`
resume path also benefits, gaining hook support and proper cleanup for
free. A new `LifecycleConfig` struct captures setup commands, timeouts,
and devcontainer phases, keeping the engine's API clean. The existing
`run()` method is unchanged, so API server and integration tests
continue working with pre-initialized sandboxes.
The `setup_remote_git` helper is moved from `cli/run.rs` into
`engine.rs` since it only depends on sandbox exec. Config is now passed
by mutable reference to `run_with_lifecycle` so the engine can fill in
remote git fields (base SHA, run branch) that downstream code needs.
Comprehensive tests verify event emission ordering, setup command
execution/failure, and cleanup behavior.
### Fabro Details
<details>
<summary>Ran 10 stages in 53m 27s for $14.42</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $6.22 | 0 |
| simplify_opus | 0s | $2.63 | 0 |
| simplify_gemini | 0s | $2.73 | 0 |
| simplify_gpt | 0s | $2.84 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **53m 27s** | **$14.42** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
74dae0a342
|
Refactor: Collapse pull_request fields on RunConfig into Option<PullRequestConfig> (#21)
This refactoring collapses four flat `pull_request_*` fields on
`RunConfig` (`pull_request_enabled`, `pull_request_draft`,
`pull_request_auto_merge`, `pull_request_merge_strategy`) into a single
`pull_request: Option<PullRequestConfig>` field. This better represents
the natural tree structure of the configuration: `None` means PR
creation is disabled, while `Some(config)` carries all PR settings
directly.
The PR creation logic in `run.rs` is restructured to use `if let
Some(ref pr_cfg) = config.pull_request` instead of checking a boolean
flag, and construction is simplified to filter out disabled configs at
build time via `.filter(|p| p.enabled).cloned()`. All test constructors
across `engine.rs`, `integration.rs`, `daytona_integration.rs`,
`server.rs`, and `manager_loop.rs` are updated from two lines
(`pull_request_enabled: false, pull_request_draft: false/true`) to a
single `pull_request: None`.
### Fabro Details
<details>
<summary>Ran 10 stages in 20m 32s for $6.62</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.71 | 0 |
| simplify_opus | 0s | $1.63 | 0 |
| simplify_gemini | 0s | $1.74 | 0 |
| simplify_gpt | 0s | $1.55 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **20m 32s** | **$6.62** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
469b5d4670
|
Fix: Stage durations always show 0s in PR descriptions (#20)
This PR fixes a bug where stage durations in PR descriptions always
displayed `0s`. The root cause was a key mismatch in
`extract_stage_durations()`: the function was building its HashMap using
`node_label` (human-readable names like `"Preflight Compile"`), but all
three lookup sites were querying by `node_id` (DOT graph identifiers
like `"preflight_compile"`). Every lookup missed and fell back to the
default value of `0`.
The fix changes the HashMap key from `node_label` to `node_id` so that
it matches what the callers actually use for lookups. Additionally, the
existing test is updated so that `node_label` and `node_id` values
differ from each other (e.g., `"plan"` vs `"Plan"`), ensuring the test
would have caught this mismatch before the fix rather than silently
passing due to both fields being identical.
### Fabro Details
<details>
<summary>Ran 10 stages in 7m 46s for $1.28</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $0.21 | 0 |
| simplify_opus | 0s | $0.35 | 0 |
| simplify_gemini | 0s | $0.35 | 0 |
| simplify_gpt | 0s | $0.38 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **7m 46s** | **$1.28** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
brynary-fabro[bot]
|
d29c1d817e
|
Wire up missing hook invocations (#19)
This PR wires up three `HookEvent` variants—`StageRetrying`,
`ParallelStart`, and `ParallelComplete`—that were defined in the enum
and documented but never actually invoked by the engine. `StageRetrying`
hooks now fire at both retry sites in `execute_with_retry` (error-retry
and explicit-Retry-status paths) immediately before the backoff sleep.
`ParallelStart` and `ParallelComplete` hooks fire in the parallel
handler after their corresponding event emissions, using a new
`EngineServices::run_hooks()` convenience method since the handler
doesn't have direct access to the engine's hook method.
The two remaining unwired events, `SandboxReady` and `SandboxCleanup`,
are marked as reserved with doc comments on the enum variants and
annotated in the docs table, since wiring them requires sandbox
lifecycle changes outside the engine's scope. A small
`HookContext::set_node()` helper is introduced to reduce repeated field
assignment across all hook call sites, and existing
`StageStart`/`StageComplete` hook calls are refactored to use it.
### Fabro Details
<details>
<summary>Ran 10 stages in 25m 6s for $5.60</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.04 | 0 |
| simplify_opus | 0s | $1.51 | 0 |
| simplify_gemini | 0s | $1.44 | 0 |
| simplify_gpt | 0s | $1.62 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **25m 6s** | **$5.60** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
brynary-fabro[bot]
|
50c032576d
|
Style SVG output from fabro graph and API (#18)
This PR styles the SVG output from `fabro graph` and the API graph
endpoints to match the polished look of the documentation SVGs. It
introduces two internal functions in `graph.rs`:
`inject_dot_style_defaults`, which inserts DOT-level defaults
(transparent background, teal `#357f9e` node strokes, gray `#666666`
edges, Helvetica font) after the first `{` in any DOT source, and
`postprocess_svg`, which removes the Graphviz-generated white background
polygon and injects a `<style>` block with `prefers-color-scheme: dark`
media queries for automatic dark mode support.
The `render_dot` function is updated to apply both transformations (DOT
defaults before rendering, SVG post-processing after), and is made `pub`
so the API layer can reuse it. Both `get_graph` in `server.rs` and
`get_run_graph` in `demo/mod.rs` are simplified from ~30 lines of inline
Graphviz process management down to a single call to a shared
`render_dot_svg` helper that delegates to `render_dot` on a blocking
thread.
Six new unit tests validate the styling pipeline: default injection with
and without braces, white background removal, dark mode style insertion,
and the existing SVG integration test is extended to assert styled
output. PNG output is unaffected by the SVG post-processing step.
### Fabro Details
<details>
<summary>Ran 10 stages in 23m 31s for $5.48</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.15 | 0 |
| simplify_opus | 0s | $1.55 | 0 |
| simplify_gemini | 0s | $1.65 | 0 |
| simplify_gpt | 0s | $1.13 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **23m 31s** | **$5.48** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
Bryan Helmkamp
|
1af5193a82
|
Handle assistant output replay on stream retries | ||
|
Bryan Helmkamp
|
f39b37da64
|
Add auto-merge support for pull requests via GitHub GraphQL API
When `auto_merge = true` is set in `[pull_request]` config, Fabro enables GitHub's auto-merge on created PRs using the `enablePullRequestAutoMerge` GraphQL mutation. Auto-merge implies `draft = false` since GitHub doesn't allow auto-merge on draft PRs. A `merge_strategy` field (squash/merge/rebase, default squash) controls the merge method. Failures to enable auto-merge (e.g. repo doesn't have the setting enabled) warn but don't fail the run. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
brynary-fabro[bot]
|
470fcfe120
|
Fix: Sub-agent file writes not tracked in API backend (#17)
This PR fixes a bug where files written by sub-agents were missing from
`outcome.files_touched`, causing downstream pipeline nodes like
`simplify_opus` to receive incomplete file lists. The root cause was
that `spawn_event_forwarder` only matched top-level
`ToolCallStarted`/`ToolCallCompleted` events, while sub-agent tool calls
arrived wrapped in `AgentEvent::SubAgentEvent` and fell through to the
`_ => {}` catch-all.
The fix extracts the file-tracking logic into a standalone
`track_file_event` function that recursively unwraps `SubAgentEvent`
layers before matching on the inner tool call events. This handles
arbitrarily nested sub-agent hierarchies (sub-sub-agents, etc.). The
three separate `Arc<Mutex<...>>` fields for pending calls, touched
files, and last file are consolidated into a single `FileTracking`
struct behind one lock, simplifying the forwarder signature and reducing
lock contention.
Four new unit tests verify the behavior: top-level write tracking,
single-level sub-agent unwrapping, double-nested sub-sub-agent
unwrapping, and proper cleanup of pending entries on tool call errors.
### Fabro Details
<details>
<summary>Ran 10 stages in 22m 1s for $4.18</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $0.91 | 0 |
| simplify_opus | 0s | $0.89 | 0 |
| simplify_gemini | 0s | $1.44 | 0 |
| simplify_gpt | 0s | $0.94 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **22m 1s** | **$4.18** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
brynary-fabro[bot]
|
cb5b7bc3a7
|
Limit command stdout/stderr to last N lines in preamble (#16)
This PR limits command stdout/stderr output in the preamble to the last
N lines, reducing token waste from verbose build progress and download
noise. Command nodes like `cargo check` or `cargo clippy` can produce
300+ lines of output, but the useful content (errors, summaries) is
almost always at the tail. Compact and summary:medium fidelity levels
now keep the last 25 lines, while summary:high keeps the last 50 lines.
Truncated output displays an `(N lines omitted)` indicator consistent
with the existing omission pattern used for stages.
The implementation adds a `tail_lines` helper that skips leading lines
beyond the limit, prepends an omission notice, and applies the
appropriate indentation to each retained line. This is applied only to
inline stdout/stderr rendering—artifact pointer branches and other
fidelity levels (summary:low, truncate, full) are intentionally
untouched. The PR includes unit tests for the `tail_lines` helper itself
as well as integration tests verifying truncation behavior at compact
and summary:high fidelity, and confirming artifact pointers remain
untruncated.
### Fabro Details
<details>
<summary>Ran 10 stages in 18m 4s for $3.45</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $0.84 | 0 |
| simplify_opus | 0s | $0.95 | 0 |
| simplify_gemini | 0s | $0.81 | 0 |
| simplify_gpt | 0s | $0.85 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **18m 4s** | **$3.45** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -q --workspace -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
brynary-fabro[bot]
|
2972e38bed
|
Add --direction flag to fabro graph (#15)
This PR adds a `--direction` (`-d`) flag to the `fabro graph` CLI
command, bringing it to parity with the web UI's LR/TB toggle buttons.
Users can now pass `--direction lr` or `--direction tb` to override the
`rankdir` declared in the DOT source before it's handed off to the `dot`
renderer.
The implementation introduces a `GraphDirection` enum (with `ValueEnum`
and `Display` derives for clap integration), an optional `--direction`
field on `GraphArgs`, and an `apply_direction` helper that uses a
lazy-compiled regex to rewrite `rankdir=…` in the DOT source—mirroring
the same approach used by the web UI. The CLI docs are updated with the
new flag, and two unit tests verify that the rewrite works correctly and
that omitting the flag leaves the source unchanged.
### Fabro Details
<details>
<summary>Ran 10 stages in 17m 6s for $3.60</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $0.68 | 0 |
| simplify_opus | 0s | $1.02 | 0 |
| simplify_gemini | 0s | $0.83 | 0 |
| simplify_gpt | 0s | $1.06 | 0 |
| verify | 0s | – | 0 |
| fmt | 0s | – | 0 |
| **Total** | **17m 6s** | **$3.60** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (13 nodes and 16
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gemini [label="Simplify (Gemini)", prompt="@prompts/simplify.md", model="gemini-3.1-pro-preview-customtools"]
simplify_gpt [label="Simplify (GPT-54)", prompt="@prompts/simplify.md", model="gpt-54"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo fmt --all 2>&1", goal_gate=true, max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gemini -> simplify_gpt -> verify
verify -> fmt [condition="outcome=success"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
Bryan Helmkamp
|
420c0874a0
|
Enable auto PR creation and add debug logging for skipped PR cases
The `[pull_request]` config in fabro.toml was missing `enabled = true`, so workflow runs silently skipped PR creation. Additionally, four skip paths in the PR creation logic had no logging at all, making it hard to diagnose why a PR wasn't opened. Added debug-level logs for: config not enabled, dry-run mode, engine error, and non-success run status. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
Bryan Helmkamp
|
b604bc4b60
|
Designate retros as experimental, disable by default via [features] flag
Move retro control from [fabro] retro to [features] retros in project config. Default changes from true to false — retros are now opt-in. Add retros field to server config Features struct, OpenAPI spec, TypeScript client, and web app config. Update docs with experimental warning and new enablement instructions. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|
brynary-fabro[bot]
|
da3610af59
|
Unified dry-run via Handler::simulate() (#14)
This PR introduces a unified dry-run mechanism by adding a
`Handler::simulate()` trait method and a `dispatch_handler()` routing
function that selects between `simulate()` and `execute()` based on
`services.dry_run`. Previously, dry-run behavior was scattered
inconsistently across handlers—`CommandHandler` checked
`services.dry_run` inline, `AgentHandler`/`PromptHandler`/`FanInHandler`
relied on the backend being `None`, and `WaitHandler`/`HumanHandler` had
no dry-run support at all (sleeping or blocking on input for real). This
made dry-run behavior fragile and difficult to extend to new handlers.
The new design adds an `Outcome::simulated(node_id)` factory for
standardized dry-run results, a default `simulate()` implementation on
the `Handler` trait that returns a generic simulated success, and
per-handler overrides where custom context updates are needed.
`CommandHandler` populates empty output/stderr, `AgentHandler` and
`PromptHandler` set simulated
`last_stage`/`last_response`/`response.{id}` context keys,
`FanInHandler` calls `heuristic_select()` without an LLM, `HumanHandler`
auto-selects the first choice, and `ParallelHandler` dispatches child
branches through `dispatch_handler()` while skipping git worktree
operations. The inline `dry_run` check in `CommandHandler::execute()` is
removed, and both call sites in the engine (`execute_with_retry` and
parallel branch dispatch) now route through `dispatch_handler()`.
All existing dry-run tests are updated to test `simulate()` directly,
and new tests verify that `dispatch_handler()` correctly routes based on
the `dry_run` flag and that each handler's `simulate()` produces the
expected context updates and outcome structure.
### Fabro Details
<details>
<summary>Ran 7 stages in 24m 57s for $4.72</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $3.29 | 0 |
| simplify | 0s | $1.42 | 0 |
| verify | 0s | – | 0 |
| **Total** | **24m 57s** | **$4.72** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Fabro Assistant <assistant@fabro.dev>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
0d7ae73857
|
Random Edge Selection (#12)
This PR introduces a `selection="random"` node attribute that enables
weighted-random tiebreaking when choosing among candidate outgoing
edges. The existing deterministic behavior (highest weight, then lexical
node ID) remains the default. The cascade priority—conditions →
preferred label → suggested next → unconditional → fallback—is
unchanged; randomness only replaces the final pick-one-from-candidates
step within each tier. A new `weighted_random` function handles the
sampling, treating edges with weight ≤ 0 as weight 1, while a
`pick_edge` dispatcher routes to either the random or deterministic
strategy based on the node's `selection()` accessor.
A validation rule (`RandomSelectionNoConditionsRule`) rejects nodes that
combine `selection="random"` with conditional edges, since condition
evaluation order would conflict with random selection. A companion rule
(`SelectionValidRule`) warns on unrecognized selection values. Both are
registered as built-in lint rules with appropriate error/warning
severities and actionable fix suggestions.
Documentation is updated in the transitions guide with a new "Random
selection" section explaining the behavior and constraints, and the DOT
language reference gains a `selection` row in the node attributes table.
All changes were developed following red/green TDD cycles with
comprehensive test coverage for the accessor, weighted random sampling,
edge selection integration, and both validation rules.
### Fabro Details
<details>
<summary>Ran 7 stages in 17m 1s for $4.07</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $2.43 | 0 |
| simplify | 0s | $1.63 | 0 |
| verify | 0s | – | 0 |
| **Total** | **17m 1s** | **$4.07** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
brynary-fabro[bot]
|
d90a805f1a
|
fabro asset CLI subcommands (#10)
This PR adds `fabro asset list` and `fabro asset cp` CLI subcommands for
inspecting and copying run artifacts (screenshots, test reports, traces)
that are collected to `~/.fabro/runs/{id}/artifacts/assets/`.
Previously, users had to manually browse the filesystem to find these
files.
The core implementation lives in a new `asset.rs` module that provides
`scan_assets()`, which walks the asset manifest files under each node's
retry directories to build structured `AssetEntry` records. The `list`
subcommand displays a formatted table (or JSON with `--json`) showing
node, retry number, size, and path for each asset. The `cp` subcommand
supports copying all assets or a specific file (via `RUN_ID:path`
syntax), with a `--tree` flag to preserve the `{node}/retry_{N}/`
directory structure and collision detection in flat mode.
The implementation reuses existing utilities throughout:
`split_run_path()` for parsing the colon-separated source syntax (made
`pub(crate)`), `resolve_run()` for run ID prefix matching,
`format_size()` for human-readable byte formatting (also made
`pub(crate)`), and `AssetCollectionSummary` for manifest
deserialization.
### Fabro Details
<details>
<summary>Ran 7 stages in 12m 31s for $3.61</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.33 | 0 |
| simplify | 0s | $2.28 | 0 |
| verify | 0s | – | 0 |
| **Total** | **12m 31s** | **$3.61** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
872ead311d
|
fabro fork subcommand (#9)
This PR adds a new `fabro fork` subcommand that creates a new run
branching from an existing run at a specific checkpoint, without
modifying the original run. This is a non-destructive alternative to
`fabro rewind` — instead of moving branch refs backward and losing later
checkpoint history, fork preserves the source run entirely and creates
fresh run and metadata branches for the new run.
The implementation heavily reuses existing infrastructure from
`rewind.rs` (timeline building, target resolution, parallel map loading,
prefix-based run ID lookup) and follows the same CLI patterns. The core
`execute_fork` function generates a new ULID, creates a run branch ref
pointing at the target checkpoint's commit, then builds a new metadata
branch containing an updated manifest (with new run ID and branch name),
the original graph, and the checkpoint state from the target commit. It
supports the same target syntax as rewind (`@N`, `node_name`,
`node_name@N`), defaults to the latest checkpoint when no target is
specified, and optionally pushes new branches to the remote.
The PR also makes `load_parallel_map` public in `rewind.rs` so fork can
reuse it, and includes five tests covering run branch creation, metadata
branch correctness, preservation of the original run, default-to-latest
behavior, and forking at a specific ordinal.
### Fabro Details
<details>
<summary>Ran 7 stages in 15m 15s for $4.39</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.94 | 0 |
| simplify | 0s | $2.44 | 0 |
| verify | 0s | – | 0 |
| **Total** | **15m 15s** | **$4.39** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
brynary-fabro[bot]
|
54dcb35080
|
Add safeguards to asset collection and checkpoint commits (#8)
This PR adds safeguards to asset collection and checkpoint commits to
prevent collecting or committing excessive files from large, untracked
directories like Python virtual environments, build outputs, and tool
caches.
Specifically, it introduces a `MAX_FILE_COUNT` limit of 100 files in
`select_files_to_collect()` to cap asset collection regardless of total
size budget, expands the `EXCLUDE_DIRS` list with seven new entries
(`.venv`, `venv`, `.cache`, `.tox`, `.pytest_cache`, `.mypy_cache`,
`dist`) to match common project directory patterns that can contain
thousands of generated files, and makes the constant public for reuse.
Notably, `build` and `env`/`.env` were intentionally omitted as too
generic or potentially conflicting with legitimate project files.
The checkpoint commit logic in `git_checkpoint()` is updated to always
apply the built-in `EXCLUDE_DIRS` as git pathspec excludes (converted to
`**/dirname/**` glob format), merged with any user-configured exclude
globs. This ensures that even with no user configuration, checkpoint
`git add -A` commands won't inadvertently stage virtual environments,
caches, or build artifacts. All changes are covered by new tests
following red/green TDD.
### Fabro Details
<details>
<summary>Ran 7 stages in 10m 34s for $2.41</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.21 | 0 |
| simplify | 0s | $1.19 | 0 |
| verify | 0s | – | 0 |
| **Total** | **10m 34s** | **$2.41** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
|
||
|
arc-1e68f1[bot]
|
43f5fb0edb
|
Inject GitHub App IAT into Sandbox as GITHUB_TOKEN (#7)
This PR adds GitHub App Installation Access Token (IAT) injection into
sandboxes, allowing `gh` CLI and other GitHub-authenticated tools to
work seamlessly inside workflow sandboxes. Workflow authors can declare
required GitHub permissions in `workflow.toml` under a `[github]`
section (e.g., `permissions = { contents = "write", pull_requests =
"read" }`), with project-wide defaults available in `fabro.toml`.
Workflow-level config fully replaces project-level defaults, consistent
with existing `[pull_request]` behavior.
The implementation introduces a `GitHubConfig` struct wired through
`WorkflowRunConfig`, `RunDefaults`, and `ProjectConfig`, with proper
`apply_defaults` (inherit if unset) and `merge_overlay` (replace if
present) semantics. At runtime, a new `mint_github_token()` helper signs
a JWT, resolves the repo's owner/repo from the origin URL, and requests
a scoped IAT which is injected as `GITHUB_TOKEN` into the sandbox
environment. The previously private
`create_installation_access_token_with_permissions` in `fabro-github` is
made public to support this. A preflight check also mints a token during
validation to surface credential or permission issues early.
Comprehensive tests cover TOML parsing with and without `[github]`,
default inheritance, workflow-over-default precedence, and overlay merge
semantics for `RunDefaults`.
### Fabro Details
<details>
<summary>Ran 7 stages in 27m 15s for $5.88</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $3.79 | 0 |
| simplify | 0s | $2.09 | 0 |
| verify | 0s | – | 0 |
| **Total** | **27m 15s** | **$5.88** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Claude <claude@anthropic.com>
|
||
|
arc-1e68f1[bot]
|
527b6252ef
|
Add goal to WorkflowRunStarted and render in fabro logs --pretty (#6)
This PR adds a `goal` field to the `WorkflowRunStarted` event so that
users can immediately see what a workflow is trying to accomplish when
reading logs. The field is an `Option<String>` with `serde(default,
skip_serializing_if)` to maintain backward compatibility with existing
JSONL logs that don't include it—mirroring the same pattern used by
`base_sha` and `run_branch`.
On the rendering side, `fabro logs --pretty` now displays the goal below
the workflow header line when present, using markdown rendering with
proper indentation and terminal-width wrapping. The markdown rendering
logic was extracted into a shared `render_indented_markdown` helper,
which is also now used by the existing `AssistantMessage` rendering to
eliminate duplication.
Tests cover round-trip serialization with a goal, backward-compatible
deserialization of old events without the field, verification that
`None` goals are omitted from JSON output, and pretty-formatting
behavior both with and without a goal present.
### Fabro Details
<details>
<summary>Ran 7 stages in 23m 38s for $3.02</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 0s | – | 0 |
| preflight_compile | 0s | – | 0 |
| preflight_lint | 0s | – | 0 |
| implement | 0s | $1.76 | 0 |
| simplify | 0s | $1.26 | 0 |
| verify | 0s | – | 0 |
| **Total** | **23m 38s** | **$3.02** | **0** |
</details>
<details>
<summary>Ran <code>ImplementAndSimplify.fabro</code> (10 nodes and 13
edges)</summary>
```dot
digraph ImplementAndSimplify {
graph [
goal="Implement and simplify",
model_stylesheet="
* { backend: api; model: claude-opus-4-6;}
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan."]
simplify [label="Simplify", prompt="@prompts/simplify.md"]
verify [label="Verify", shape=parallelogram, script="cargo clippy -- -D warnings 2>&1 && cargo test 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings and test failures.", max_visits=3]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=success"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=success"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=success"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify -> verify
verify -> exit [condition="outcome=success"]
verify -> fixup
fixup -> verify
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
||
|
arc-1e68f1[bot]
|
b61786ff30
|
Rename metadata branch from refs/fabro/{run_id} to fabro/meta/{run_id} (#3)
* fabro(01KKS6PG9929P116A2RRKXC738): toolchain (success) Fabro-Run: 01KKS6PG9929P116A2RRKXC738 Fabro-Completed: 2 Fabro-Checkpoint: |