Two fabro-server scenarios run their container on the catalog image
ghcr.io/lithoscomputer/ubuntu-22.04:slim and wait about five seconds
for the run to finish; on a runner without the image the plugin's pull
takes longer than that. Pull it with the default runner image before
the suite, and give the Docker job the default runner image pull too,
since its fabro-petri Docker test runs that image.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every Petri run takes its scope through a sandbox-driver plugin
executable that Petri finds on PATH, so the test jobs need
sandbox-driver-host and sandbox-driver-docker installed at the rev the
workspace pins. The three jobs share one from-source install through an
actions/cache entry keyed on the OS and the rev.
The Linux test job also pre-pulls Petri's default runner image, which
the suite's Docker scenarios leave to Petri: the plugin pulls it on
first use, but a 1 GiB pull inside a run's timeout is a flake.
The stdio plugin job was built for the deleted fabro-sandbox layer. It
becomes the Docker providers job: the `docker_` scenario variants and
the fabro-petri suite, with the fabro-sandbox and fabro-workflow steps
whose tests no longer exist removed.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every intra-doc link `cargo doc --workspace --no-deps` warned on now
resolves or is plain code: the private constant and helper, the removed
`InterpString::resolve`, the lithos `Message`, the sandbox-driver facets,
the `RunOptions::git_author` the cutover removed, and the stale
`platform_record_for` paragraph on the platform records. The `[@REF]`
segment of `fabro run`'s help text is allowed as help, not a link. A
`Rustdoc` job runs the same command with `-D warnings`.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Fabro runs its workflows on Petri. The six Petri packages and the testkit
are pinned by revision in the workspace manifest under `petri_*` keys, and
`fabro-petri` is the one crate that depends on them. The crate's tests run
the `hello` bundle in memory on the stub registry and a command-only
workflow on the host sandbox; both skip without the sandbox-driver host
plugin, and the sandbox-plugins CI job requires it.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The step's filter named `fabro-agent` and `fabro-llm`. This branch removes
`fabro-agent`, and `fabro-llm` has no ignored tests, so the step ran nothing
and nextest exited 4 on the empty selection. The agent loop's tests run in
the ordinary suite now and pebble's own suite covers the loop.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Replace fabro's hand-written agent loop with pebble's `CodingAgent` and
delete the `fabro-agent` crate.
Workflow: `PebbleBackend` builds one agent per stage over `RunSandbox`,
binds the stage's hooks as tool middleware, the interviewer as the
human-input provider, and a durable `EventSink` that writes every agent
event through the run event log before the agent goes on. Full-fidelity
threads continue across stages through `export`/`resume_from_export`.
Model failover takes the session record after the failed prompt and
continues it on the next route with `ResumeMode::UseModel`, so no tool
effect repeats. The steering hub targets pebble's control handle, with
a steering lease holding completion open while a human is paired.
Events: `EventBody::Agent` carries pebble's `CodingAgentEvent` envelope;
the per-variant bodies, the transcript projection, and the fabro-only
context-window, tool-summary, and skill types are gone in favor of
pebble's. The OpenAPI schemas, generated Rust and TypeScript clients,
and web readers follow.
Ask Fabro: the session runs a `CodingAgent` under a read-only permission
policy and a system prompt transform. Its conversation lives in a new
`run_session_records` table and resumes on the recorded model with the
event cursor advanced past the run log.
`fabro exec` builds the same agent over a local sandbox with pebble's
permission middleware and an interactive approval service.
The catalog fills in `metadata.agent.profile` for operator providers
that declare none, so pebble's lookup is the one resolution path.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Both sides rewrote the same crates. This branch replaced fabro's sandbox
layer with the sandbox driver: one RunSandbox, no Sandbox trait, driver
events consumed directly, MockSandbox over the driver's doubles. Main
replaced fabro's LLM layer with lithos-llm: fabro-model deleted, the
catalog and provider ids from lithos, credentials through the lithos
CredentialProvider, clients built with build_client.
Every conflict was one of those two renames meeting in an import list or
a signature, so the rule was mechanical: sandbox names resolve to this
branch, LLM names to main. Where main's newer code still used the old
sandbox API — new session tests over Arc::new(MockSandbox), the SDK
example's LocalSandbox, test fakes typed as Arc<dyn Sandbox> — it is
ported to RunSandbox and the mock helper. Where this branch still used
fabro-model or Client::from_source, main's replacement stands. One
combined future in the CLI runner crossed clippy's size budget and is
boxed at its call.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The plugin scenarios launched two executables fabro built itself,
`fabro-sandbox-host` and `fabro-sandbox-docker`, that only wrapped the
driver's providers in a stdio server the driver already ships as
`sandbox-driver-host` and `sandbox-driver-docker`. Fabro now finds the
driver's executables on PATH: CI installs them at the rev the workspace
pins, read from Cargo.toml so the plugins and the in-process providers
are one build, and a developer installs them the same way. The plugin
proof in fabro-sandbox skips without the executable unless the CI
environment forbids skipping; it was also never running in CI, which
ran it under `--run-ignored only` although it is not ignored, so the
job now runs it on its own.
The pin moves to the head of the sandbox-driver PR stack #9 through
#15: tag pins, classified git failures, the stop grace ladder, snapshot
ensure, the ownership scope, and the testing doubles, which the next
commits adopt. The `sandbox-driver-testing` crate joins the workspace
dependencies for them.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Closes the sandbox-driver adoption: any provider a sandbox-driver plugin
executable serves can now host a fabro run, and fabro's own bundled
providers can be served the same way.
- `SandboxSpec::Plugin` builds a normalized driver spec from the
environment (image or Dockerfile source, or a provider-managed
directory; resources; network policy; labels; env) and lays fabro's
repository checkout out inside the provider's working directory. The
layout is recorded on the run through the new `workspace_layout` trait
method.
- Plugin settings on a bundled kind (`[server.sandbox.providers.docker]
path = ...`) serve that kind out of process through the driver's
executable; the config layer no longer rejects them.
- `ProviderAccess` carries the server's provider settings and the vault's
Daytona credentials to every reconnect: run resume, sandbox details,
terminals, previews, and the worker's start path. The worker receives
the settings through `StartServices`. No "plugin not wired" errors
remain.
- The CLI worker requires GitHub credentials only when a repository will
be cloned; a `none` target on a clone-based provider creates an empty
workspace and needs none.
- fabro-db tracks its migrations directory so a new migration file
recompiles the crate; the environment provider migration had been
silently missing from stale builds. Environment store 500s now log
their cause.
- The CLI workflow scenarios run against `host-plugin` (the driver's
Host executable under the non-bundled `host` kind) and `docker-plugin`
(the bundled `docker` kind served over stdio), each on an isolated
server, printing the server log on failure. A live Daytona gate runs
the native git clone over the JSON-RPC wire. A new CI job runs the
plugin scenarios and the driver-backed Docker integration tests with
the plugin executables built.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Add lithos-llm as a pinned git dependency and replace the in-repo
`test/twin/openai` crate with the published `twins` revision that
lithos-llm verifies its codecs against. Move the Fabro policy overlay
(`fabro-policy.toml`) into fabro-llm so Fabro owns its own catalog
policy layer.
Drop the twin-openai nextest overrides and CI package filter now that the
crate is no longer a workspace member.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Rust 1.98.0 (released 2026-08-20) passes --fix-cortex-a53-843419 to the
linker for aarch64-unknown-linux-musl, which the zig cc wrapper used by
cargo-zigbuild rejects, breaking the release build for that target. Pin
all workflows that installed unpinned stable to 1.97.1 until the zig
toolchain handles the new flag. The nightly-2026-04-14 fmt/clippy
toolchains are unchanged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
## What
Fixes the `openai_twin_*` parity-matrix failures that have been on
`main` since #449: every multi-turn scenario whose scripted response
includes text fails on its second turn with 400 `"message input items
require supported content"`.
## Root cause
Two twin behaviors collided (bisected: passes at #447, fails at #449):
1. **The twin's streaming `response.output_item.done` for message items
omitted the `content` array** (`test/twin/openai/src/sse.rs`) — it sent
only `id`/`type`/`status`/`role`, where the real API sends the completed
item in full. The openai adapter preserves message output items verbatim
(`ContentPart::Other { kind: OPENAI_MESSAGE }`) and replays them as
assistant history on the next turn — required so reasoning items keep
their "required following item" in Responses round-trips. So the replay
arrived content-less.
2. **#449 tightened the twin's input validation** to also validate
explicit `type: "message"` items (previously only type-less items were
validated as messages; anything with an explicit type was accepted
unchecked). The twin started rejecting its own round-tripped output.
The new validation caught a real infidelity in the emitter — the emit
side is what's wrong.
Nobody noticed because **CI never runs the twin e2e suites**: `rust.yml`
runs `--profile ci` without `--run-ignored`, so the parity matrix only
runs when someone invokes the e2e profile locally.
## Fix
- The streamed message `output_item.done` now carries its `output_text`
content, matching the real API and the twin's own non-streaming
`responses_json()`.
- The input validator accepts `output_text` parts on **assistant**
message items (the real API allows these; the twin's non-streaming
responses already require it for faithful replay). Non-assistant
`output_text` parts get a dedicated rejection message.
## Tests
- New contract test
`responses_stream_message_item_done_round_trips_as_input`: streams a
response, asserts the completed message item carries its `output_text`
content, and replays the item verbatim as assistant-history input,
asserting the twin accepts its own output.
- `cargo nextest run -p twin-openai` — 56 passed
- `cargo nextest run -p fabro-agent -E 'test(parity)' --run-ignored
only` — **91/91 passed** (was 7 failing)
- `cargo nextest run -p fabro-llm --run-ignored only` — 10 passed
- `cargo nextest run --workspace` — green apart from two pre-existing
env-dependent `fabro-workflow` failures that reproduce on clean `main`
in shells with provider API keys exported (unrelated; CI is green on
them because it has no such keys)
- clippy `-D warnings` / fmt — clean
Found while reviewing #481 (whose parity runs surfaced this); #481
itself is unaffected — it doesn't touch the openai adapter or the twin,
and the failures exist on its merge-base.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
## CI (separate commit, drop if unwanted)
`ci: run twin-mode e2e suites on Linux` adds a step to the existing
Linux test job running the ignored twin-mode suites for the packages
that are fully green today (`fabro-agent`, `fabro-llm`, `twin-openai`) —
104 tests, ~1s on a warm build, no secrets needed (live-only tests
self-skip in twin mode). This is what would have caught the #449
regression. The remaining ignored suites (fabro-cli twin tests,
Docker/Daytona sandbox tests, fabro-spa asset test) need their own fixes
before joining; widen the `-E` filter as they're cleaned up. Note the
step deliberately avoids the `e2e` nextest profile, since
`NEXTEST_PROFILE=e2e` implies strict mode, which fails on missing
secrets.
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
## Summary
Enforce Cargo lockfile use across CI and release automation so jobs fail
on stale `Cargo.lock` state instead of resolving dependencies
implicitly. This adds `--locked` to Rust CI, release builds/tests,
nightly release tagging, the TypeScript workflow's embedded Rust build,
and helper-owned Cargo calls in `fabro-dev`.
This also fixes the Linux CI flake exposed by the PR: canceling a
durably blocked in-process run could take the abort path while the
workflow was still unwinding a human-input gate, causing
`run.failed(cancelled)` to be followed by `run.unblocked`. That invalid
event order broke projection rebuilds and made `GET /runs/{id}` return
404. Cancellation now uses the durable lifecycle status when selecting
the in-process blocked-run path, so the pending interview is cancelled
before the terminal event is emitted.
The release command's intentional `cargo update --workspace` step is
unchanged, because that step updates `Cargo.lock` after bumping the
workspace version.
## Testing
- `cargo nextest run --locked -p fabro-dev --features dev -E
'test(dry_run_computes_stable_version_from_date) |
test(dry_run_prints_equivalent_build_commands)'`
- `cargo --locked dev release --dry-run --skip-tests --release-date
2026-01-01`
- `cargo --locked dev docs check`
- `cargo nextest run --locked -p fabro-server --features test-support
cancel_durably_blocked_in_process_run_cancels_pending_interview_without_abort_signal
--status-level fail --final-status-level fail --show-progress none`
- `cargo nextest run --locked -p fabro-server --features test-support
--test it scenario::lifecycle --profile ci --status-level fail
--final-status-level fail --show-progress none --no-fail-fast`
- Linux Docker stress reproduction: `cargo nextest run --locked -p
fabro-server --features test-support --test it
scenario::lifecycle::full_http_lifecycle_cancel --profile ci
--stress-count 200 --status-level fail --final-status-level fail
--show-progress none`
- `cargo +nightly-2026-04-14 fmt --check --all`
- `cargo +nightly-2026-04-14 clippy --locked -p fabro-dev --features dev
--all-targets -- -D warnings`
- `cargo +nightly-2026-04-14 clippy --locked -p fabro-server --features
test-support --all-targets -- -D warnings`
- `git diff --check`
Full `cargo nextest run --locked -p fabro-dev --features dev` currently
has two unrelated policy-test failures:
`policy::catalog_builtin_references_stay_in_allowlist` and
`policy::workflow_template_rendering_call_sites_stay_in_allowlist`.
---
[](https://github.com/EveryInc/compound-engineering-plugin)
🤖 Generated with GPT-5 via [Codex](https://openai.com/codex)
---------
Co-authored-by: Release Repro <release-repro@example.com>
Invert the docs convention so the Mintlify-published site lives under
docs/public/ and internal artifacts (strategy docs, brainstorms, plans,
etc.) sit at docs/ root or docs/internal/. Tools that default to writing
into docs/ now land in the catch-all instead of leaking into the
published tree.
- Move Mintlify content (administration/, agents/, api-reference/,
changelog/, core-concepts/, examples/, execution/, getting-started/,
human-tools/, integrations/, languages/, reference/, tutorials/,
workflows/, images/, logo/, docs.json, favicon.svg, dot-highlight.js)
into docs/public/.
- Collapse docs-internal/ into docs/internal/.
- Update Rust path references (fabro-api/build.rs, fabro-server,
fabro-dev), TypeScript generator arg, CI path filters, clippy.toml
reasons, AGENTS.md/CLAUDE.md, and README.md image refs.
Mintlify dashboard project root must be updated to docs/public/ in a
follow-up. .mintignore move/trim and .claude/skills/ updates land in a
separate commit.
Add a settings reference generator backed by OptionsMetadata on the sparse config layer structs. The generated user-configuration page is fenced and checked in CI alongside the CLI reference.
Add a cargo dev generator for the CLI reference and gate the generated docs in CI. The generator reads the fabro clap command tree through a narrow public reference surface so CLI docs drift is caught without exposing runtime command internals.
clippy.toml already bans std::env::{set_var,remove_var} via
disallowed_methods, and every existing call site carries a scoped
#[expect(clippy::disallowed_methods, reason = "...")]. The shell grep
is redundant and forced a second, less granular allowlist.
Also update server-secrets-strategy.md to describe clippy as the
enforcement mechanism.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Move server-only settings reads out of user-facing CLI commands into a
dedicated local_server module, the install/uninstall exceptions, and the
worker subcommand. Adds bin/dev/check-boundary.sh to prevent regressions.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Stacked cleanup of the `canonicalize blocked run status` work (local
commit `d13cdf374`) plus reconciliation with origin's `canonicalize
paginated run list responses` (origin commit `8ab689da7`). Both efforts
ran in parallel and diverged on the column name (`blocked` vs `waiting`)
and on how the board response is shaped — this PR converges them,
keeping `blocked` as the canonical column id while adopting origin's
`column` field on `RunListItem` and `StoreRunSummary` shape.
Also fixes a production-worker regression introduced by the
canonicalization: the worker's start-precondition only accepted
`Submitted | Starting`, so once runs started transitioning through
`Queued` on the way to `Starting`, every subprocess-worker run failed
with `Precondition failed: cannot start run: status is Queued`. That
cascaded into ~90 failing CLI/server integration tests locally.
## Commits
1. `f65843168` refactor(runs): simplify blocked status follow-ups
2. `1492d956c` chore: resolve clippy warnings
3. `676fd9f44` first merge of origin/main
4. `23fc92a2f` **fix(runs): allow Queued status in start precondition**
← the cascade-fix
5. `36b507a83` refactor: simplify pause/unpause + dedupe web status
tables
6. `8d8d27748` refactor(workflow): encapsulate BlockedStateTracker
inside HumanHandler
7. `1c17fda35` second merge of origin/main — resolves waiting vs blocked
8. `4cd3ef7b1` refactor(workflow): Mutex<usize> → AtomicUsize
9. `2e5a58e8a` fix(demo): align run-4 lifecycle status with Blocked
board column
## Test plan
- [x] fmt, clippy, build, doctests all clean
- [x] `cargo nextest run --workspace` — **4092/4092 pass**
- [x] `bun test` — **26/26 pass**, typecheck + production build clean
- [x] Manual CLI repro of the Queued-precondition fix
- [x] Browser smoke test: all 5 columns render with correct
labels/colors, demo run-4 appears in Blocked lane with question text
intact
## Known follow-up (not blocking)
A "paused-while-blocked" run (status `Paused` + `blocked_reason: Some`)
lands in the `running` column because the visible status chooses
`Paused` over `Blocked`. The pending question is not prominent on the
board. Addressing it would require `board_column()` to branch on
`(status, blocked_reason)` rather than just `status` — worth a separate
ticket.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Start every workflow with permissions: {} and grant the minimum
required per job, following Astral's defense-in-depth pattern so a
newly added job can't silently inherit repo read access.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a ci nextest profile (30s slow-timeout, terminate-after 4, 2s leak-
timeout) and wires rust.yml's test and test-macos jobs to use it. Local
invocations keep using the tight default profile.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Default nextest emits a PASS line per test, which scrolls thousands of
lines in CI. `slow` shows only slow/failing tests plus the summary.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
actions/checkout v4 → v6.0.2, actions/upload-artifact v4 → v7.0.1,
actions/download-artifact v4 → v8.0.1. Silences the Node 20 deprecation
warnings ahead of the June 2026 forced cutover. All pins are full
commit SHAs with version comments.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- rust.yml: move clippy to nightly-2026-04-14 (was stable); also pin
fmt to the same nightly date for consistency. Both jobs now use the
dated nightly and the run-step uses `cargo +nightly-2026-04-14 ...`.
- AGENTS.md: update developer commands to match CI.
- Duration constructors: replace `Duration::from_secs(N * 60)` /
`Duration::from_millis(N * 1000)` with `from_mins` / `from_secs` /
`from_hours` across the workspace to satisfy clippy's new
`duration_suboptimal_units` lint. std::time::Duration only — custom
`settings::duration::Duration` sites kept on `from_secs`.
- map/unwrap_or cleanup: `.map(f).unwrap_or(v)` → `.map_or(v, f)`,
`.map(f).unwrap_or(false)` on Result → `.is_ok_and(f)`, per
`clippy::map_unwrap_or`.
- Misc lints: collapse nested `if` into match guard in
handler/llm/api.rs and run_state.rs; replace `columns.len() > 0`
with `!columns.is_empty()`; switch a pair of `sort_by` calls to
`sort_by_key`.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Addresses supply chain hardening items from #160:
- Pin taiki-e/install-action to commit SHA
- Pin rust:1-bookworm and oven/bun:1 to image digests
- Pin mintlify to 4.2.507
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Move async subprocess paths to Tokio or spawn_blocking, document the
intentional synchronous std::process::Command callsites, and make CI run
Clippy with --all-targets so the guardrail applies to test code too.
## Summary
- **Unify foreground and detach code paths**: Both `fabro run` modes now
go through the same `create_run() + start_run()` pipeline, with
foreground adding `attach_run()`. Only `--preflight` remains as a
special case.
- **Fix three bugs in create→start→attach path**: (1) `_run_engine`
crashed for `.fabro` workflows by hardcoding `run.toml` — now falls back
to `graph.fabro`; (2) `attach_run` couldn't detect crashed engines due
to zombie processes — `start_run` now returns the `Child` handle; (3)
`create_run` ignored `--run-id`.
- **Configure nextest slow-timeout profiles**: Tighten unit test timeout
to 2s slow / 4s kill, add `e2e` profile with 10s/30s. Switch CI and docs
to `cargo nextest run`.
## Test plan
- [ ] `cargo nextest run --workspace` passes with new timeout profiles
- [ ] `fabro run <workflow>` works in foreground mode (create + start +
attach)
- [ ] `fabro run --detach <workflow>` prints run ID and exits
- [ ] `fabro attach <run>` works standalone (without child handle)
- [ ] `fabro resume <run>` works for both `.toml` and `.fabro` workflows
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Remove rust-toolchain/rust-cache setup (pre-installed on self-hosted),
disable checkout clean, enable sccache and incremental compilation.
Split macOS tests into a separate job that only runs on workflow_dispatch
since release.yml already gates on macOS tests.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add release.yml: tag-triggered (v*) workflow that cross-compiles
for aarch64-apple-darwin and x86_64-unknown-linux-gnu, packages
archives with SHA256 checksums, and creates a GitHub Release
- Add workflow_dispatch, concurrency groups, and cache-on-failure
to rust.yml and typescript.yml
- Run clippy and test on both ubuntu-latest and macos-15
- Add release profile with thin LTO and strip to Cargo.toml
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
ubuntu-latest-16-cores requires a GitHub Team/Enterprise org plan and
was causing all jobs to stay queued indefinitely.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Updated: Cargo.toml workspace members, CI workflow paths, Dockerfile
COPY, AGENTS.md, skill mapping, and doc references.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Rust workflow runs fmt, clippy, and test on crate/cargo/openapi changes.
TypeScript workflow runs typecheck, test, and build on apps/packages changes.
Both use path filters so changes don't cross-trigger.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
