diff --git a/run.json b/run.json
index a59fc14a1..266e2877f 100644
--- a/run.json
+++ b/run.json
@@ -505,7 +505,7 @@
     "kind": "running"
   },
   "status_updated_at": "2026-07-09T23:18:01.698576530Z",
-  "last_event_at": "2026-07-09T23:23:40.959911074Z",
+  "last_event_at": "2026-07-09T23:51:28.527418314Z",
   "pending_control": null,
   "checkpoints": [
     {
@@ -789,9 +789,9 @@
       }
     },
     {
-      "seq": 0,
+      "seq": 66,
       "checkpoint": {
-        "timestamp": "2026-07-09T23:23:41.052289967Z",
+        "timestamp": "2026-07-09T23:23:44.868754737Z",
         "current_node": "implement",
         "completed_nodes": [
           "start",
@@ -802,27 +802,148 @@
         ],
         "node_retries": {},
         "context_values": {
-          "internal.run_id": "01KX4JRJ2NXPQASFNE8PPKTQX6",
           "failure_class": "deterministic",
           "internal.retry_count.start": 0,
-          "internal.retry_count.preflight_compile": 0,
-          "current_node": "implement",
-          "failure_signature": "implement|deterministic|api_deterministic|openai|authentication",
           "graph.rankdir": "LR",
-          "graph.model_stylesheet": "\n            * { model: claude-opus-4-8; }\n        ",
-          "thread.preflight_lint.current_node": "implement",
+          "outcome": "failed",
+          "thread.toolchain.current_node": "preflight_compile",
+          "failure_signature": "implement|deterministic|api_deterministic|openai|authentication",
           "graph.goal": "# Demote `server.integrations.slack.default_channel` to a plain literal string\n\n**Self-contained implementation plan.** Everything needed to implement this is\nin this file plus the repository. Independent — no preconditions; can land\nanytime. (It must land before a separate, later effort freezes a registry of\nconfig-field kinds, but nothing in this plan depends on that.)\n\n> **Token notation.** Interpolation tokens are written in this file without\n> their enclosing double curly braces, so the file is safe to pass directly as\n> a workflow goal (the goal templater would otherwise try to expand them).\n> Read `env.NAME`, `vars.NAME`, `secrets.NAME` as the double-curly-brace token\n> form used in the codebase, and write the real double-brace syntax in the\n> code, tests, and docs you produce.\n\n## Context and goal\n\n`server.integrations.slack.default_channel` is the last server-defined config\nfield still typed `InterpString`. Every other server-scope field was demoted\nto a plain literal under the project's rule that interpolation belongs to\nfields resolved with run context — server-startup consumption is served\nnatively by shells/compose/systemd, and a token there just ferries an env var\nacross a process boundary.\n\nThis field is exactly that case:\n\n- It was **born `Option<String>`** (2026-03-05, in the original Slack\n  integration crate; adopted into server settings 2026-04-07) and was\n  documented from the start with literal examples only. The v2 settings\n  schema typed it `InterpString` (2026-04-09) and wired env resolution the\n  same day as part of that schema's uniform staged design — not a\n  field-specific feature, and never requested or documented as interpolable.\n  The one place env-interpolated Slack channels ARE a deliberate, documented\n  feature is run-scope notification routes (added 2026-05-23) — a surface\n  this change does not touch. The uniform-staging capability was superseded\n  by the later interpolation-taxonomy decision that startup-consumed server\n  fields stay literal, under which every comparable server field was already\n  demoted.\n- It resolves **once, at server startup, env-only** — no secrets, no vars,\n  and never re-resolved at message-send time.\n- **No documentation ever advertised interpolation** on it; every doc example\n  is a plain literal channel name.\n- The **run-scope interpolating surface already exists** and is the\n  user-facing one: `run.notifications.<route>.slack.channel` and\n  `run.interviews.slack.channel` are `InterpString` with variable\n  substitution at run creation. The server field is only the zero-config\n  fallback destination for interview prompts.\n- Product direction reinforces it: in hosted deployments users have no access\n  to server env at all (their surface is variables and secrets at run scope),\n  and a future chat-integration plugin system should inherit a simple literal\n  field, not a special-case interpolating one.\n\n**Goal:** change the field to a plain `Option<String>` end-to-end, drop the\nstartup resolution, and emit the standard demoted-field warning when a\nleftover token-shaped value is found — matching how the earlier server-field\ndemotions were shipped.\n\nDesign rules (fixed):\n\n- **Keep the field** as the operator's literal fallback. Do not remove it or\n  relocate it; run scope already covers per-run needs.\n- **Wire-invisible.** `InterpString` serializes as its raw source string, so\n  the stored/wire JSON shape is unchanged by this demote. The OpenAPI schema\n  already models the field as a plain string — no spec change.\n- **Warn, don't break.** A value still containing a token-shaped span (double\n  curly braces) parses fine as a literal; emit the existing demoted-field\n  warning at resolve time so the ~3 months of nightly builds where an env\n  token would have resolved get a loud, non-fatal migration signal.\n\n## Verified current state (as of main `d5dcd1179`, 2026-07-09 — re-verify before starting; line numbers are anchors, not gospel)\n\n- Type: `lib/crates/fabro-types/src/settings/server.rs:258-271` —\n  `SlackIntegrationSettings { enabled: bool, default_channel:\n  Option<InterpString> }`.\n- Config layer: `lib/crates/fabro-config/src/layers/server.rs:237` —\n  `default_channel: Option<InterpString>`.\n- Resolve copy-through: `lib/crates/fabro-config/src/resolve/server.rs:365-369`\n  (clones the field into resolved server settings).\n- Startup resolution: `lib/crates/fabro-server/src/server.rs:2422-2430` —\n  `slack_settings.default_channel.as_ref().map(|value|\n  value.resolve(process_env_var)...)` feeding\n  `SlackService::new(bot, app, default_channel: Option<String>)`\n  (`server.rs:589-600`). The service posts interview prompts to it\n  (`server.rs:~657` `let Some(default_channel) = ... else return`, `:689`\n  `post_message`).\n- Demoted-field warning helper: `warn_if_demoted_template` in\n  `lib/crates/fabro-config/src/resolve/` (see its callers in\n  `resolve/cli.rs:50-76` for the exact usage pattern: field path string +\n  `Option<&str>` value).\n- Run-scope channels (untouched by this PR):\n  `run.notifications.<route>` slack channel and\n  `run.interviews.slack.channel`, both `InterpString`, variable-substituted at\n  run creation (`fabro-types/src/settings/run.rs`, `substitute_variables`).\n- Docs mentioning the field (all literal examples):\n  `docs/public/administration/server-configuration.mdx:453`,\n  `docs/public/human-tools/interviews.mdx:97`,\n  `docs/public/integrations/slack.mdx:112-117`.\n- OpenAPI: `docs/public/api-reference/fabro-api.yaml:13619-13623` models\n  `default_channel` as a nullable plain string in the relevant schema —\n  expected to need **no change**.\n\n## Implementation\n\n1. **Type change**: `fabro-types/src/settings/server.rs` and\n   `fabro-config/src/layers/server.rs` — `Option<InterpString>` →\n   `Option<String>`. Chase the compiler through the resolve copy-through and\n   any settings merge/serde helpers.\n2. **Demotion warning**: at the server-settings resolve site, call the\n   existing demoted-field warning helper with the field path\n   `server.integrations.slack.default_channel` and the literal value,\n   following the exact pattern of its existing callers. The warning must log\n   the field path and guidance only — never treat the value as sensitive\n   output beyond what the existing helper does.\n3. **Drop the startup resolve**: `fabro-server/src/server.rs` — pass the\n   literal through to `SlackService::new` directly; delete the\n   `resolve(process_env_var)` call and its error mapping. `SlackService`\n   itself is unchanged (it already takes `Option<String>`).\n4. **Verify (read-only) the interview routing preference**: confirm whether\n   the interview-prompt posting path prefers `run.interviews.slack.channel`\n   over the server default when both are set. If it does not, **do not build\n   routing changes** — record the finding in the PR description as a\n   follow-up observation.\n5. **Docs**: the three pages above already show literals; adjust wording only\n   if any implies interpolation (none is expected to). If the generated\n   options reference annotates the field's type, regenerate via\n   `cargo dev docs` and confirm `cargo dev docs check` is green.\n\n## Scope boundaries — deliberately NOT in this PR\n\n- **Run-scope notification and interview channel fields** — leave as-is; they\n  are the intended interpolating surface and are already correct.\n- **Slack credential resolution** (bot/app tokens via the vault at startup) —\n  leave as-is; unrelated to the channel field.\n- **Interview prompt routing behavior** — observe and report only (step 4);\n  changing which channel wins is separate product work.\n- **Any chat-integration plugin restructuring** — future work; this PR only\n  simplifies what that redesign will inherit.\n- **The config-field kind registry and its conformance tests** — separate\n  planned work; do not start it here.\n\nIf work outside these boundaries seems genuinely required for this PR to\ncompile or pass its tests, stop and state that in the PR description rather\nthan expanding scope.\n\n## Tests (failing-first; hermetic — no ambient env dependence)\n\n- A literal channel (`#releases`) parses, merges, and reaches\n  `SlackService::new` unchanged.\n- A value containing a token-shaped span parses as a **literal** (no\n  resolution, no error) and emits the demoted-field warning naming\n  `server.integrations.slack.default_channel` (log-capture, matching how the\n  existing demotion-warning tests assert).\n- Startup wiring: with the field set, the service receives exactly the\n  configured string; with it absent, `None` (existing behavior preserved).\n- Existing Slack/server integration tests pass unchanged.\n- Serde shape: a settings round-trip of the field produces the same JSON/TOML\n  as before the type change (wire-invisibility).\n\n## Acceptance / verification\n\n- `cargo +nightly-2026-04-14 fmt --check --all`\n- `cargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings`\n- `cargo nextest run --workspace`\n- `cargo dev docs check`\n- No OpenAPI/wire change (assert: `docs/public/api-reference/fabro-api.yaml`\n  untouched; TypeScript client regeneration not needed).\n\n## Conventions\n\n- Plain-English commit messages, PR text, and comments — describe what the\n  change does; no internal planning identifiers or plan-file names in\n  anything that ships.\n- PR description must state plainly: the field only ever resolved env vars,\n  only at server startup, and was never documented as interpolable; anyone\n  with a token in the field gets a startup warning and the value is treated\n  as a literal; per-run channels (`run.notifications`, `run.interviews.slack`)\n  remain the interpolating surface; no wire change.\n",
           "internal.fidelity": "compact",
+          "internal.retry_count.preflight_lint": 0,
+          "internal.retry_count.toolchain": 0,
+          "internal.work_dir": "/home/daytona/workspace/fabro",
+          "internal.run_id": "01KX4JRJ2NXPQASFNE8PPKTQX6",
+          "internal.node_visit_count": 1,
           "internal.retry_count.implement": 0,
           "thread.preflight_compile.current_node": "preflight_lint",
-          "internal.work_dir": "/home/daytona/workspace/fabro",
-          "internal.thread_id": "preflight_lint",
+          "thread.preflight_lint.current_node": "implement",
           "command.output": "blob://sha256/12ae32cb1ec02d01eda3581b127c1fee3b0dc53572ed6baf239721a03d82e126",
-          "internal.node_visit_count": 1,
+          "graph.model_stylesheet": "\n            * { model: claude-opus-4-8; }\n        ",
+          "internal.thread_id": "preflight_lint",
+          "internal.retry_count.preflight_compile": 0,
+          "current_node": "implement",
+          "thread.start.current_node": "toolchain"
+        },
+        "node_outcomes": {
+          "preflight_lint": {
+            "status": "succeeded",
+            "context_updates": {
+              "command.output": "blob://sha256/12ae32cb1ec02d01eda3581b127c1fee3b0dc53572ed6baf239721a03d82e126"
+            },
+            "notes": "Script completed: cargo +nightly-2026-04-14 clippy -q --workspace --all-targets -- -D warnings 2>&1",
+            "usage": null,
+            "timing": {
+              "wall_time_ms": 0,
+              "inference_time_ms": 0,
+              "tool_time_ms": 162141,
+              "active_time_ms": 162141
+            }
+          },
+          "start": {
+            "status": "succeeded",
+            "usage": null
+          },
+          "implement": {
+            "status": "failed",
+            "failure": {
+              "message": "LLM error: Authentication error for openai: Your authentication token has been invalidated. Please try signing in again.",
+              "category": "deterministic",
+              "signature": "api_deterministic|openai|authentication"
+            },
+            "usage": null
+          },
+          "preflight_compile": {
+            "status": "succeeded",
+            "context_updates": {
+              "command.output": "blob://sha256/12ae32cb1ec02d01eda3581b127c1fee3b0dc53572ed6baf239721a03d82e126"
+            },
+            "notes": "Script completed: cargo check -q --workspace 2>&1",
+            "usage": null,
+            "timing": {
+              "wall_time_ms": 0,
+              "inference_time_ms": 0,
+              "tool_time_ms": 161483,
+              "active_time_ms": 161483
+            }
+          },
+          "toolchain": {
+            "status": "succeeded",
+            "context_updates": {
+              "command.output": "blob://sha256/20eeffec02497fbda7b51f51b06fe29c1d639551eee4d5ea9845fc1f86bd77e1"
+            },
+            "notes": "Script completed: 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",
+            "usage": null,
+            "timing": {
+              "wall_time_ms": 0,
+              "inference_time_ms": 0,
+              "tool_time_ms": 1284,
+              "active_time_ms": 1284
+            }
+          }
+        },
+        "next_node_id": "simplify_fable",
+        "git_commit_sha": "f5706f0c2462949fb44070405dfdfe6ca89d07ab",
+        "loop_failure_signatures": {
+          "implement|deterministic|api_deterministic|openai|authentication": 1
+        },
+        "node_visits": {
+          "preflight_compile": 1,
+          "preflight_lint": 1,
+          "implement": 1,
+          "start": 1,
+          "toolchain": 1
+        }
+      },
+      "diff": {
+        "summary": {
+          "files_changed": 0,
+          "additions": 0,
+          "deletions": 0
+        }
+      }
+    },
+    {
+      "seq": 0,
+      "checkpoint": {
+        "timestamp": "2026-07-09T23:51:28.657657781Z",
+        "current_node": "simplify_fable",
+        "completed_nodes": [
+          "start",
+          "toolchain",
+          "preflight_compile",
+          "preflight_lint",
+          "implement",
+          "simplify_fable"
+        ],
+        "node_retries": {},
+        "context_values": {
+          "internal.run_id": "01KX4JRJ2NXPQASFNE8PPKTQX6",
+          "failure_class": "",
+          "internal.retry_count.start": 0,
+          "thread.implement.current_node": "simplify_fable",
+          "failure_signature": "",
+          "graph.model_stylesheet": "\n            * { model: claude-opus-4-8; }\n        ",
+          "graph.goal": "# Demote `server.integrations.slack.default_channel` to a plain literal string\n\n**Self-contained implementation plan.** Everything needed to implement this is\nin this file plus the repository. Independent — no preconditions; can land\nanytime. (It must land before a separate, later effort freezes a registry of\nconfig-field kinds, but nothing in this plan depends on that.)\n\n> **Token notation.** Interpolation tokens are written in this file without\n> their enclosing double curly braces, so the file is safe to pass directly as\n> a workflow goal (the goal templater would otherwise try to expand them).\n> Read `env.NAME`, `vars.NAME`, `secrets.NAME` as the double-curly-brace token\n> form used in the codebase, and write the real double-brace syntax in the\n> code, tests, and docs you produce.\n\n## Context and goal\n\n`server.integrations.slack.default_channel` is the last server-defined config\nfield still typed `InterpString`. Every other server-scope field was demoted\nto a plain literal under the project's rule that interpolation belongs to\nfields resolved with run context — server-startup consumption is served\nnatively by shells/compose/systemd, and a token there just ferries an env var\nacross a process boundary.\n\nThis field is exactly that case:\n\n- It was **born `Option<String>`** (2026-03-05, in the original Slack\n  integration crate; adopted into server settings 2026-04-07) and was\n  documented from the start with literal examples only. The v2 settings\n  schema typed it `InterpString` (2026-04-09) and wired env resolution the\n  same day as part of that schema's uniform staged design — not a\n  field-specific feature, and never requested or documented as interpolable.\n  The one place env-interpolated Slack channels ARE a deliberate, documented\n  feature is run-scope notification routes (added 2026-05-23) — a surface\n  this change does not touch. The uniform-staging capability was superseded\n  by the later interpolation-taxonomy decision that startup-consumed server\n  fields stay literal, under which every comparable server field was already\n  demoted.\n- It resolves **once, at server startup, env-only** — no secrets, no vars,\n  and never re-resolved at message-send time.\n- **No documentation ever advertised interpolation** on it; every doc example\n  is a plain literal channel name.\n- The **run-scope interpolating surface already exists** and is the\n  user-facing one: `run.notifications.<route>.slack.channel` and\n  `run.interviews.slack.channel` are `InterpString` with variable\n  substitution at run creation. The server field is only the zero-config\n  fallback destination for interview prompts.\n- Product direction reinforces it: in hosted deployments users have no access\n  to server env at all (their surface is variables and secrets at run scope),\n  and a future chat-integration plugin system should inherit a simple literal\n  field, not a special-case interpolating one.\n\n**Goal:** change the field to a plain `Option<String>` end-to-end, drop the\nstartup resolution, and emit the standard demoted-field warning when a\nleftover token-shaped value is found — matching how the earlier server-field\ndemotions were shipped.\n\nDesign rules (fixed):\n\n- **Keep the field** as the operator's literal fallback. Do not remove it or\n  relocate it; run scope already covers per-run needs.\n- **Wire-invisible.** `InterpString` serializes as its raw source string, so\n  the stored/wire JSON shape is unchanged by this demote. The OpenAPI schema\n  already models the field as a plain string — no spec change.\n- **Warn, don't break.** A value still containing a token-shaped span (double\n  curly braces) parses fine as a literal; emit the existing demoted-field\n  warning at resolve time so the ~3 months of nightly builds where an env\n  token would have resolved get a loud, non-fatal migration signal.\n\n## Verified current state (as of main `d5dcd1179`, 2026-07-09 — re-verify before starting; line numbers are anchors, not gospel)\n\n- Type: `lib/crates/fabro-types/src/settings/server.rs:258-271` —\n  `SlackIntegrationSettings { enabled: bool, default_channel:\n  Option<InterpString> }`.\n- Config layer: `lib/crates/fabro-config/src/layers/server.rs:237` —\n  `default_channel: Option<InterpString>`.\n- Resolve copy-through: `lib/crates/fabro-config/src/resolve/server.rs:365-369`\n  (clones the field into resolved server settings).\n- Startup resolution: `lib/crates/fabro-server/src/server.rs:2422-2430` —\n  `slack_settings.default_channel.as_ref().map(|value|\n  value.resolve(process_env_var)...)` feeding\n  `SlackService::new(bot, app, default_channel: Option<String>)`\n  (`server.rs:589-600`). The service posts interview prompts to it\n  (`server.rs:~657` `let Some(default_channel) = ... else return`, `:689`\n  `post_message`).\n- Demoted-field warning helper: `warn_if_demoted_template` in\n  `lib/crates/fabro-config/src/resolve/` (see its callers in\n  `resolve/cli.rs:50-76` for the exact usage pattern: field path string +\n  `Option<&str>` value).\n- Run-scope channels (untouched by this PR):\n  `run.notifications.<route>` slack channel and\n  `run.interviews.slack.channel`, both `InterpString`, variable-substituted at\n  run creation (`fabro-types/src/settings/run.rs`, `substitute_variables`).\n- Docs mentioning the field (all literal examples):\n  `docs/public/administration/server-configuration.mdx:453`,\n  `docs/public/human-tools/interviews.mdx:97`,\n  `docs/public/integrations/slack.mdx:112-117`.\n- OpenAPI: `docs/public/api-reference/fabro-api.yaml:13619-13623` models\n  `default_channel` as a nullable plain string in the relevant schema —\n  expected to need **no change**.\n\n## Implementation\n\n1. **Type change**: `fabro-types/src/settings/server.rs` and\n   `fabro-config/src/layers/server.rs` — `Option<InterpString>` →\n   `Option<String>`. Chase the compiler through the resolve copy-through and\n   any settings merge/serde helpers.\n2. **Demotion warning**: at the server-settings resolve site, call the\n   existing demoted-field warning helper with the field path\n   `server.integrations.slack.default_channel` and the literal value,\n   following the exact pattern of its existing callers. The warning must log\n   the field path and guidance only — never treat the value as sensitive\n   output beyond what the existing helper does.\n3. **Drop the startup resolve**: `fabro-server/src/server.rs` — pass the\n   literal through to `SlackService::new` directly; delete the\n   `resolve(process_env_var)` call and its error mapping. `SlackService`\n   itself is unchanged (it already takes `Option<String>`).\n4. **Verify (read-only) the interview routing preference**: confirm whether\n   the interview-prompt posting path prefers `run.interviews.slack.channel`\n   over the server default when both are set. If it does not, **do not build\n   routing changes** — record the finding in the PR description as a\n   follow-up observation.\n5. **Docs**: the three pages above already show literals; adjust wording only\n   if any implies interpolation (none is expected to). If the generated\n   options reference annotates the field's type, regenerate via\n   `cargo dev docs` and confirm `cargo dev docs check` is green.\n\n## Scope boundaries — deliberately NOT in this PR\n\n- **Run-scope notification and interview channel fields** — leave as-is; they\n  are the intended interpolating surface and are already correct.\n- **Slack credential resolution** (bot/app tokens via the vault at startup) —\n  leave as-is; unrelated to the channel field.\n- **Interview prompt routing behavior** — observe and report only (step 4);\n  changing which channel wins is separate product work.\n- **Any chat-integration plugin restructuring** — future work; this PR only\n  simplifies what that redesign will inherit.\n- **The config-field kind registry and its conformance tests** — separate\n  planned work; do not start it here.\n\nIf work outside these boundaries seems genuinely required for this PR to\ncompile or pass its tests, stop and state that in the PR description rather\nthan expanding scope.\n\n## Tests (failing-first; hermetic — no ambient env dependence)\n\n- A literal channel (`#releases`) parses, merges, and reaches\n  `SlackService::new` unchanged.\n- A value containing a token-shaped span parses as a **literal** (no\n  resolution, no error) and emits the demoted-field warning naming\n  `server.integrations.slack.default_channel` (log-capture, matching how the\n  existing demotion-warning tests assert).\n- Startup wiring: with the field set, the service receives exactly the\n  configured string; with it absent, `None` (existing behavior preserved).\n- Existing Slack/server integration tests pass unchanged.\n- Serde shape: a settings round-trip of the field produces the same JSON/TOML\n  as before the type change (wire-invisibility).\n\n## Acceptance / verification\n\n- `cargo +nightly-2026-04-14 fmt --check --all`\n- `cargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings`\n- `cargo nextest run --workspace`\n- `cargo dev docs check`\n- No OpenAPI/wire change (assert: `docs/public/api-reference/fabro-api.yaml`\n  untouched; TypeScript client regeneration not needed).\n\n## Conventions\n\n- Plain-English commit messages, PR text, and comments — describe what the\n  change does; no internal planning identifiers or plan-file names in\n  anything that ships.\n- PR description must state plainly: the field only ever resolved env vars,\n  only at server startup, and was never documented as interpolable; anyone\n  with a token in the field gets a startup warning and the value is treated\n  as a literal; per-run channels (`run.notifications`, `run.interviews.slack`)\n  remain the interpolating surface; no wire change.\n",
+          "internal.fidelity": "compact",
+          "last_response": "## Summary\n\nThe prior \"implement\" stage had failed on an API authentication error with **zero changes committed**, so I implemented the plan from scratch, then ran the three-agent simplify review and ",
+          "internal.work_dir": "/home/daytona/workspace/fabro",
+          "internal.thread_id": "implement",
           "thread.toolchain.current_node": "preflight_compile",
           "internal.retry_count.preflight_lint": 0,
-          "outcome": "failed",
+          "last_stage": "simplify_fable",
+          "response.simplify_fable": "## Summary\n\nThe prior \"implement\" stage had failed on an API authentication error with **zero changes committed**, so I implemented the plan from scratch, then ran the three-agent simplify review and applied its findings.\n\n### Implementation\n- **Type demotion**: `default_channel: Option<InterpString>` → `Option<String>` in `fabro-types/src/settings/server.rs:261` and `fabro-config/src/layers/server.rs:237` (unused `InterpString` import removed from fabro-types).\n- **Demotion warning**: `resolve_integrations` in `fabro-config/src/resolve/server.rs` now calls `warn_if_demoted_template(\"server.integrations.slack.default_channel\", ...)`, mirroring the adjacent GitHub arm.\n- **Dropped startup resolution**: `fabro-server/src/server.rs:2425` passes the literal straight to `SlackService::new`; the `resolve(process_env_var)` call and its error mapping are gone. Deleted the now-single-caller `display_interp` helper in `server/handler/system.rs` — the `/system/integrations` endpoint no longer does per-request template resolution.\n- **Tests**: literal pass-through + JSON wire-shape lock, token-stays-literal (mirroring `resolve_storage_root_keeps_template_token_literal`), and startup wiring tests (set → verbatim `#releases`; absent → `None`).\n\n### Step-4 finding (read-only, for the PR description)\nThe interview-prompt posting path (`SlackService::handle_event`, server.rs:657) posts **only** to the server `default_channel` and never consults `run.interviews.slack.channel` — that run-scope field is parsed and resolved but has no consumer in the posting path. Follow-up product work, not changed here. Also worth noting in the PR: previously an unresolvable env token **failed startup**; now it flows through verbatim with a resolve-time warning — the intended \"warn, don't break\" behavior.\n\n### Simplify review results\n- **Fixed**: extracted `slack_test_vault_tokens()` to replace three verbatim copies of the vault-token fixture; aligned `system.rs` to the sibling `as_ref()` + clone-at-insert pattern.\n- **Skipped (with reason)**: dropping the JSON wire-shape assert (plan explicitly requires the serde wire-invisibility test) and folding the two wiring tests into one (plan requires both the set and absent cases).\n- Reviewers confirmed clean: no dead code, correct helper reuse, field-path string style matches siblings, private-field asserts match module test style, and the diff is a net perf win (startup and per-request resolution both removed).\n\n### Verification\n- `cargo +nightly-2026-04-14 fmt --check --all` — clean\n- `cargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings` — clean\n- `cargo nextest run --workspace` — 6961 passed (before final test-file cleanup; fabro-server/fabro-config re-run after: 984 passed)\n- `cargo dev docs check` — up to date; the three doc pages already show literals only\n- `docs/public/api-reference/fabro-api.yaml` — untouched; no wire change, no TS client regen needed",
+          "internal.retry_count.preflight_compile": 0,
+          "thread.preflight_lint.current_node": "implement",
+          "internal.retry_count.implement": 0,
+          "current_node": "simplify_fable",
+          "graph.rankdir": "LR",
+          "thread.preflight_compile.current_node": "preflight_lint",
+          "command.output": "blob://sha256/12ae32cb1ec02d01eda3581b127c1fee3b0dc53572ed6baf239721a03d82e126",
+          "internal.node_visit_count": 1,
+          "outcome": "succeeded",
           "thread.start.current_node": "toolchain",
+          "internal.retry_count.simplify_fable": 0,
           "internal.retry_count.toolchain": 0
         },
         "node_outcomes": {
@@ -840,6 +961,53 @@
               "active_time_ms": 161483
             }
           },
+          "simplify_fable": {
+            "status": "succeeded",
+            "context_updates": {
+              "response.simplify_fable": "## Summary\n\nThe prior \"implement\" stage had failed on an API authentication error with **zero changes committed**, so I implemented the plan from scratch, then ran the three-agent simplify review and applied its findings.\n\n### Implementation\n- **Type demotion**: `default_channel: Option<InterpString>` → `Option<String>` in `fabro-types/src/settings/server.rs:261` and `fabro-config/src/layers/server.rs:237` (unused `InterpString` import removed from fabro-types).\n- **Demotion warning**: `resolve_integrations` in `fabro-config/src/resolve/server.rs` now calls `warn_if_demoted_template(\"server.integrations.slack.default_channel\", ...)`, mirroring the adjacent GitHub arm.\n- **Dropped startup resolution**: `fabro-server/src/server.rs:2425` passes the literal straight to `SlackService::new`; the `resolve(process_env_var)` call and its error mapping are gone. Deleted the now-single-caller `display_interp` helper in `server/handler/system.rs` — the `/system/integrations` endpoint no longer does per-request template resolution.\n- **Tests**: literal pass-through + JSON wire-shape lock, token-stays-literal (mirroring `resolve_storage_root_keeps_template_token_literal`), and startup wiring tests (set → verbatim `#releases`; absent → `None`).\n\n### Step-4 finding (read-only, for the PR description)\nThe interview-prompt posting path (`SlackService::handle_event`, server.rs:657) posts **only** to the server `default_channel` and never consults `run.interviews.slack.channel` — that run-scope field is parsed and resolved but has no consumer in the posting path. Follow-up product work, not changed here. Also worth noting in the PR: previously an unresolvable env token **failed startup**; now it flows through verbatim with a resolve-time warning — the intended \"warn, don't break\" behavior.\n\n### Simplify review results\n- **Fixed**: extracted `slack_test_vault_tokens()` to replace three verbatim copies of the vault-token fixture; aligned `system.rs` to the sibling `as_ref()` + clone-at-insert pattern.\n- **Skipped (with reason)**: dropping the JSON wire-shape assert (plan explicitly requires the serde wire-invisibility test) and folding the two wiring tests into one (plan requires both the set and absent cases).\n- Reviewers confirmed clean: no dead code, correct helper reuse, field-path string style matches siblings, private-field asserts match module test style, and the diff is a net perf win (startup and per-request resolution both removed).\n\n### Verification\n- `cargo +nightly-2026-04-14 fmt --check --all` — clean\n- `cargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings` — clean\n- `cargo nextest run --workspace` — 6961 passed (before final test-file cleanup; fabro-server/fabro-config re-run after: 984 passed)\n- `cargo dev docs check` — up to date; the three doc pages already show literals only\n- `docs/public/api-reference/fabro-api.yaml` — untouched; no wire change, no TS client regen needed",
+              "last_response": "## Summary\n\nThe prior \"implement\" stage had failed on an API authentication error with **zero changes committed**, so I implemented the plan from scratch, then ran the three-agent simplify review and ",
+              "last_stage": "simplify_fable"
+            },
+            "notes": "Stage completed: simplify_fable",
+            "usage": {
+              "input": {
+                "usage": {
+                  "model": {
+                    "provider": "anthropic",
+                    "model_id": "claude-fable-5"
+                  },
+                  "tokens": {
+                    "input_tokens": 86411,
+                    "output_tokens": 31302,
+                    "reasoning_tokens": 0,
+                    "cache_read_tokens": 3391207,
+                    "cache_write_tokens": 342310
+                  }
+                },
+                "facts": {
+                  "algorithm": "anthropic",
+                  "cache_write_5m_tokens": 342310,
+                  "cache_write_1h_tokens": 0
+                }
+              },
+              "total_usd_micros": 10099292
+            },
+            "files_touched": [
+              "/home/daytona/workspace/fabro/lib/crates/fabro-config/src/layers/server.rs",
+              "/home/daytona/workspace/fabro/lib/crates/fabro-config/src/resolve/server.rs",
+              "/home/daytona/workspace/fabro/lib/crates/fabro-config/src/tests/resolve_server.rs",
+              "/home/daytona/workspace/fabro/lib/crates/fabro-server/src/server.rs",
+              "/home/daytona/workspace/fabro/lib/crates/fabro-server/src/server/handler/system.rs",
+              "/home/daytona/workspace/fabro/lib/crates/fabro-server/src/server/tests.rs",
+              "/home/daytona/workspace/fabro/lib/crates/fabro-types/src/settings/server.rs"
+            ],
+            "timing": {
+              "wall_time_ms": 0,
+              "inference_time_ms": 621318,
+              "tool_time_ms": 1041791,
+              "active_time_ms": 1663109
+            }
+          },
           "start": {
             "status": "succeeded",
             "usage": null
@@ -882,11 +1050,12 @@
             }
           }
         },
-        "next_node_id": "simplify_fable",
+        "next_node_id": "simplify_gpt",
         "node_visits": {
           "start": 1,
           "preflight_lint": 1,
           "toolchain": 1,
+          "simplify_fable": 1,
           "implement": 1,
           "preflight_compile": 1
         }
@@ -920,6 +1089,335 @@
   "superseded_by": null,
   "pending_interviews": {},
   "stages": {
+    "simplify_fable@1": {
+      "first_event_seq": 69,
+      "prompt": null,
+      "response": null,
+      "completion": null,
+      "provider_used": {
+        "mode": "agent",
+        "provider": "anthropic",
+        "model": "claude-fable-5",
+        "reasoning_effort": "xhigh"
+      },
+      "diff": null,
+      "script_invocation": null,
+      "script_timing": null,
+      "parallel_results": null,
+      "output": null,
+      "started_at": "2026-07-09T23:23:44.870909672Z",
+      "handler": "agent",
+      "usage": {
+        "input_tokens": 86411,
+        "output_tokens": 31302,
+        "total_tokens": 3851230,
+        "reasoning_tokens": 0,
+        "cache_read_tokens": 3391207,
+        "cache_write_tokens": 342310,
+        "total_usd_micros": 10099292
+      },
+      "model": {
+        "provider": "anthropic",
+        "model_id": "claude-fable-5"
+      },
+      "todos": {
+        "kind": "anthropic_tasks",
+        "list_id": "anthropic_tasks:3bebbd10-1a84-4b50-8840-c5d5f4550480",
+        "items": [
+          {
+            "id": "1",
+            "status": "completed",
+            "order": 0,
+            "subject": "Implement demote of slack default_channel to plain String",
+            "description": "Change server.integrations.slack.default_channel from Option<InterpString> to Option<String> in fabro-types and fabro-config, add demoted-field warning at resolve time, drop startup resolution in fabro-server.",
+            "active_form": "Implementing the type demotion"
+          },
+          {
+            "id": "2",
+            "status": "completed",
+            "order": 1,
+            "subject": "Add/adjust tests for demoted field",
+            "description": "Tests: literal parses/merges/reaches SlackService unchanged; token-shaped value parses as literal and warns; serde round-trip unchanged; existing tests pass.",
+            "active_form": "Writing tests"
+          },
+          {
+            "id": "3",
+            "status": "completed",
+            "order": 2,
+            "subject": "Verify interview routing preference (read-only)",
+            "description": "Confirm whether interview-prompt posting prefers run.interviews.slack.channel over server default when both set. Record finding only.",
+            "active_form": "Verifying interview routing"
+          },
+          {
+            "id": "4",
+            "status": "completed",
+            "order": 3,
+            "subject": "Docs check and full verification",
+            "description": "Check three docs pages, run cargo dev docs check, fmt, clippy, nextest; assert no OpenAPI change.",
+            "active_form": "Running verification"
+          },
+          {
+            "id": "5",
+            "status": "completed",
+            "order": 4,
+            "subject": "Simplify review pass (3 parallel agents)",
+            "description": "Launch code-reuse, code-quality, and efficiency review agents over the final diff; fix findings.",
+            "active_form": "Running simplify review"
+          }
+        ]
+      },
+      "subagents": [
+        {
+          "agent_id": "6a22a61f",
+          "depth": 1,
+          "task": "You are a code-reuse reviewer for the Rust repository at /home/daytona/workspace/fabro. The full diff under review is saved at /tmp/demote-diff.txt — read it first.\n\nContext: this diff demotes the config field `server.integrations.slack.default_channel` from `Option<InterpString>` to `Option<String>` end-to-end, adds a demoted-field warning via the existing `warn_if_demoted_template` helper in lib/crates/fabro-config/src/resolve/mod.rs, drops a startup `resolve()` call in fabro-server/src/server.rs, and adds tests. This mirrors prior demotion PRs (git commits accd91a0a and 2bd04c793).\n\nYour job (review only, do NOT edit files):\n1. For each change in the diff, search the codebase (grep tool or rg) for existing utilities/helpers that could replace newly written code.\n2. Flag any new function that duplicates existing functionality; suggest the existing one.\n3. Flag any inline logic that could use an existing utility — string manipulation, path handling, ad-hoc checks, test fixture setup that duplicates existing helpers.\n4. Check the new tests in lib/crates/fabro-config/src/tests/resolve_server.rs and lib/crates/fabro-server/src/server/tests.rs for duplication with existing test helpers or existing tests that could be extended instead.\n\nThis is a greenfield-style codebase; favor maximum simplicity.\n\nReport a concise list of findings with file:line references, each with severity (high/medium/low) and a concrete suggested fix, or state \"no findings\" if clean.",
+          "status": {
+            "kind": "completed",
+            "success": true,
+            "turns_used": 23
+          }
+        },
+        {
+          "agent_id": "b1354a67",
+          "depth": 1,
+          "task": "You are a code-quality reviewer for the Rust repository at /home/daytona/workspace/fabro. The full diff under review is saved at /tmp/demote-diff.txt — read it first.\n\nContext: this diff demotes the config field `server.integrations.slack.default_channel` from `Option<InterpString>` to `Option<String>` end-to-end, adds a demoted-field warning via the existing `warn_if_demoted_template` helper in lib/crates/fabro-config/src/resolve/mod.rs, drops a startup `resolve()` call in fabro-server/src/server.rs, and adds tests. It intentionally mirrors prior demotion PRs (git commits accd91a0a and 2bd04c793) — check consistency with their patterns.\n\nYour job (review only, do NOT edit files):\n1. Redundant state: any state duplicating existing state, values that could be derived.\n2. Parameter sprawl: new parameters added where restructuring would be cleaner.\n3. Copy-paste with slight variation: near-duplicate blocks that should be unified (pay attention to the new `map_or` closure in lib/crates/fabro-config/src/resolve/server.rs resolve_integrations — compare with how the github arm in the same function structures its warn calls; also compare the new server test with the existing slack tests in fabro-server/src/server/tests.rs).\n4. Leaky abstractions: exposing internals that should be encapsulated (e.g., the new test asserts on the private field `service.default_channel` — check whether that's consistent with existing test style in that module).\n5. Stringly-typed code: raw strings where constants/enums exist (e.g., the field-path string \"server.integrations.slack.default_channel\" — check how sibling call sites write field paths).\n6. Check for leftover dead code: anything the diff should have deleted but didn't (e.g., unused imports, helpers with no remaining callers such as display_interp in fabro-server/src/server/handler/system.rs — verify it was fully removed and nothing else became dead).\n\nReport a concise list of findings with file:line references, each with severity (high/medium/low) and a concrete suggested fix, or state \"no findings\" if clean.",
+          "status": {
+            "kind": "completed",
+            "success": true,
+            "turns_used": 48
+          }
+        },
+        {
+          "agent_id": "2b826658",
+          "depth": 1,
+          "task": "You are an efficiency reviewer for the Rust repository at /home/daytona/workspace/fabro. The full diff under review is saved at /tmp/demote-diff.txt — read it first.\n\nContext: this diff demotes the config field `server.integrations.slack.default_channel` from `Option<InterpString>` to `Option<String>` end-to-end, adds a demoted-field warning via the existing `warn_if_demoted_template` helper in lib/crates/fabro-config/src/resolve/mod.rs (read that function), drops a startup `resolve()` call in fabro-server/src/server.rs, and adds tests.\n\nYour job (review only, do NOT edit files):\n1. Unnecessary work: redundant computations, unnecessary clones (check the `.clone()` calls added in lib/crates/fabro-config/src/resolve/server.rs and fabro-server/src/server.rs and fabro-server/src/server/handler/system.rs:~188-195 — is `clone()` vs borrowing appropriate given ownership needs? compare with sibling code).\n2. Hot-path bloat: the resolve and warning code runs at server startup and settings resolution; the system handler code runs per API request — check nothing expensive was added (e.g., does warn_if_demoted_template parse the value every time? is that consistent with its other call sites?).\n3. Missed concurrency: not likely relevant, but note anything sequential that should be parallel.\n4. Memory: unbounded structures, leaks — unlikely here, but check.\n5. Overly broad test setup: do the new tests build more app state than needed compared to sibling tests?\n\nReport a concise list of findings with file:line references, each with severity (high/medium/low) and a concrete suggested fix, or state \"no findings\" if clean.",
+          "status": {
+            "kind": "completed",
+            "success": true,
+            "turns_used": 14
+          }
+        }
+      ],
+      "skills": {
+        "available": [
+          {
+            "name": "rust-style-guide",
+            "description": "Apply this Rust style guide when writing, reviewing, refactoring, or configuring Rust code for this project. Covers Rust 2024/MSRV, library vs application conventions, public API design, errors, panics, ownership and cloning, async/Tokio/concurrency, tracing, rustfmt/Clippy, testing with nextest, and unsafe/macro policy. Also use when setting up new Rust projects, investigating Rust performance, verifying library releases, or reviewing Rust code changes."
+          }
+        ],
+        "activated": []
+      },
+      "permission_level": "full",
+      "agent_tools": [
+        {
+          "name": "AskUserQuestion",
+          "description": "Ask the human one or more questions and wait for their answers before continuing this stage.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": false
+        },
+        {
+          "name": "TaskCreate",
+          "description": "Create pending tasks in the current session. Use concise subjects, descriptions, optional activeForm text, and metadata. Check TaskList first to avoid duplicate tasks.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": true
+        },
+        {
+          "name": "TaskGet",
+          "description": "Get one task by taskId, including subject, status, description, owner, blockedBy, and blocks.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": false
+        },
+        {
+          "name": "TaskList",
+          "description": "List tasks for the current session, including status, owner, and blocking dependencies. Use TaskGet with a taskId for full description and dependency details.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": false
+        },
+        {
+          "name": "TaskUpdate",
+          "description": "Update an existing task's status, text, owner, metadata, or dependencies. Valid statuses are pending, in_progress, completed, and deleted. After completing a task, call TaskList to find newly unblocked work.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": true
+        },
+        {
+          "name": "close_agent",
+          "description": "Close a running subagent that is no longer needed.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "subagent",
+          "invoked": false
+        },
+        {
+          "name": "edit_file",
+          "description": "Edit a file by replacing an exact string. The old_string must be an exact match and unique unless replace_all is true; include surrounding context when needed. Read the file first and preserve existing indentation.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "write",
+          "invoked": true
+        },
+        {
+          "name": "glob",
+          "description": "Find files by file names using a glob pattern. Use path to choose the search root. Prefer this over shell find or ls when locating repository files.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "read",
+          "invoked": false
+        },
+        {
+          "name": "grep",
+          "description": "Search file contents with a regex pattern. Use path to choose the search root, glob_filter to limit matching files, case_insensitive for case folding, and max_results to cap output.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "read",
+          "invoked": true
+        },
+        {
+          "name": "read_file",
+          "description": "Read files before editing them. Returns line-numbered text and supports offset/limit for large files. Use this instead of shell cat, head, tail, or sed when inspecting repository files.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "read",
+          "invoked": true
+        },
+        {
+          "name": "send_input",
+          "description": "Send a follow-up message to a running subagent when new information or corrected instructions are needed.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "subagent",
+          "invoked": false
+        },
+        {
+          "name": "shell",
+          "description": "Execute shell commands for terminal operations, package managers, tests and builds. Use dedicated tools for file reads, file edits, filename searches, and content searches. Provide timeout_ms for long-running commands.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "shell",
+          "invoked": true
+        },
+        {
+          "name": "spawn_agent",
+          "description": "Spawn a subagent for independent work or context isolation. Use it for tasks that can proceed separately, and avoid duplicating the same work in the parent session.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "subagent",
+          "invoked": true
+        },
+        {
+          "name": "use_skill",
+          "description": "Load a skill's instructions by name. Call this when the user's request matches an available skill.",
+          "source": {
+            "kind": "skill"
+          },
+          "category": "other",
+          "invoked": false
+        },
+        {
+          "name": "wait",
+          "description": "Wait for a subagent to complete, then use the result to synthesize the outcome for the user.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "subagent",
+          "invoked": true
+        },
+        {
+          "name": "web_fetch",
+          "description": "Fetch content from a URL that starts with http:// or https://. Pass a prompt to extract specific information or summarize the page; omit prompt to return the page content.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": false
+        },
+        {
+          "name": "web_search",
+          "description": "Search the web using Brave Search when current external information is needed. Returns result titles, URLs, and descriptions; use web_fetch for a specific URL.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "other",
+          "invoked": false
+        },
+        {
+          "name": "write_file",
+          "description": "Create new files, or overwrite an existing file only when replacement is explicitly intended. Prefer edit_file for targeted changes to existing files because write_file overwrites the full file content.",
+          "source": {
+            "kind": "native"
+          },
+          "category": "write",
+          "invoked": false
+        }
+      ],
+      "context_window": {
+        "provider": "anthropic",
+        "model": "claude-fable-5",
+        "context_window_tokens": 1000000,
+        "input_tokens": 99056,
+        "usage_percent": 9.9056,
+        "count_method": "response_usage_scaled_breakdown",
+        "staleness": "live",
+        "generated_at": "2026-07-09T23:51:28.526564137Z",
+        "event_seq": 557,
+        "breakdown": [
+          {
+            "category": "system_prompt",
+            "tokens": 2262,
+            "usage_percent": 0.2262
+          },
+          {
+            "category": "tools",
+            "tokens": 2581,
+            "usage_percent": 0.2581
+          },
+          {
+            "category": "skills",
+            "tokens": 299,
+            "usage_percent": 0.0299
+          },
+          {
+            "category": "memory",
+            "tokens": 5487,
+            "usage_percent": 0.5487
+          },
+          {
+            "category": "conversation",
+            "tokens": 88416,
+            "usage_percent": 8.8416
+          },
+          {
+            "category": "other",
+            "tokens": 11,
+            "usage_percent": 0.0011
+          }
+        ],
+        "warnings": []
+      },
+      "state": "running"
+    },
     "start@1": {
       "first_event_seq": 18,
       "prompt": null,
@@ -1054,7 +1552,12 @@
       "first_event_seq": 52,
       "prompt": null,
       "response": null,
-      "completion": null,
+      "completion": {
+        "outcome": "failed",
+        "notes": null,
+        "failure_reason": "LLM error: Authentication error for openai: Your authentication token has been invalidated. Please try signing in again.",
+        "timestamp": "2026-07-09T23:23:41.051667955Z"
+      },
       "provider_used": {
         "mode": "agent",
         "provider": "openai",
@@ -1068,6 +1571,12 @@
       "output": null,
       "started_at": "2026-07-09T23:23:40.540288281Z",
       "handler": "agent",
+      "timing": {
+        "wall_time_ms": 510,
+        "inference_time_ms": 0,
+        "tool_time_ms": 0,
+        "active_time_ms": 0
+      },
       "usage": {
         "input_tokens": 0,
         "output_tokens": 0,
@@ -1223,7 +1732,7 @@
           "invoked": false
         }
       ],
-      "state": "running"
+      "state": "failed"
     },
     "toolchain@1": {
       "first_event_seq": 22,
diff --git a/stages/005-implement@1/status.json b/stages/005-implement@1/status.json
new file mode 100644
index 000000000..4e7838784
--- /dev/null
+++ b/stages/005-implement@1/status.json
@@ -0,0 +1,6 @@
+{
+  "outcome": "failed",
+  "notes": null,
+  "failure_reason": "LLM error: Authentication error for openai: Your authentication token has been invalidated. Please try signing in again.",
+  "timestamp": "2026-07-09T23:23:41.051667955Z"
+}
\ No newline at end of file
diff --git a/stages/006-simplify_fable@1/prompt.md b/stages/006-simplify_fable@1/prompt.md
new file mode 100644
index 000000000..7d93c0c5a
--- /dev/null
+++ b/stages/006-simplify_fable@1/prompt.md
@@ -0,0 +1,248 @@
+Goal: # Demote `server.integrations.slack.default_channel` to a plain literal string
+
+**Self-contained implementation plan.** Everything needed to implement this is
+in this file plus the repository. Independent — no preconditions; can land
+anytime. (It must land before a separate, later effort freezes a registry of
+config-field kinds, but nothing in this plan depends on that.)
+
+> **Token notation.** Interpolation tokens are written in this file without
+> their enclosing double curly braces, so the file is safe to pass directly as
+> a workflow goal (the goal templater would otherwise try to expand them).
+> Read `env.NAME`, `vars.NAME`, `secrets.NAME` as the double-curly-brace token
+> form used in the codebase, and write the real double-brace syntax in the
+> code, tests, and docs you produce.
+
+## Context and goal
+
+`server.integrations.slack.default_channel` is the last server-defined config
+field still typed `InterpString`. Every other server-scope field was demoted
+to a plain literal under the project's rule that interpolation belongs to
+fields resolved with run context — server-startup consumption is served
+natively by shells/compose/systemd, and a token there just ferries an env var
+across a process boundary.
+
+This field is exactly that case:
+
+- It was **born `Option<String>`** (2026-03-05, in the original Slack
+  integration crate; adopted into server settings 2026-04-07) and was
+  documented from the start with literal examples only. The v2 settings
+  schema typed it `InterpString` (2026-04-09) and wired env resolution the
+  same day as part of that schema's uniform staged design — not a
+  field-specific feature, and never requested or documented as interpolable.
+  The one place env-interpolated Slack channels ARE a deliberate, documented
+  feature is run-scope notification routes (added 2026-05-23) — a surface
+  this change does not touch. The uniform-staging capability was superseded
+  by the later interpolation-taxonomy decision that startup-consumed server
+  fields stay literal, under which every comparable server field was already
+  demoted.
+- It resolves **once, at server startup, env-only** — no secrets, no vars,
+  and never re-resolved at message-send time.
+- **No documentation ever advertised interpolation** on it; every doc example
+  is a plain literal channel name.
+- The **run-scope interpolating surface already exists** and is the
+  user-facing one: `run.notifications.<route>.slack.channel` and
+  `run.interviews.slack.channel` are `InterpString` with variable
+  substitution at run creation. The server field is only the zero-config
+  fallback destination for interview prompts.
+- Product direction reinforces it: in hosted deployments users have no access
+  to server env at all (their surface is variables and secrets at run scope),
+  and a future chat-integration plugin system should inherit a simple literal
+  field, not a special-case interpolating one.
+
+**Goal:** change the field to a plain `Option<String>` end-to-end, drop the
+startup resolution, and emit the standard demoted-field warning when a
+leftover token-shaped value is found — matching how the earlier server-field
+demotions were shipped.
+
+Design rules (fixed):
+
+- **Keep the field** as the operator's literal fallback. Do not remove it or
+  relocate it; run scope already covers per-run needs.
+- **Wire-invisible.** `InterpString` serializes as its raw source string, so
+  the stored/wire JSON shape is unchanged by this demote. The OpenAPI schema
+  already models the field as a plain string — no spec change.
+- **Warn, don't break.** A value still containing a token-shaped span (double
+  curly braces) parses fine as a literal; emit the existing demoted-field
+  warning at resolve time so the ~3 months of nightly builds where an env
+  token would have resolved get a loud, non-fatal migration signal.
+
+## Verified current state (as of main `d5dcd1179`, 2026-07-09 — re-verify before starting; line numbers are anchors, not gospel)
+
+- Type: `lib/crates/fabro-types/src/settings/server.rs:258-271` —
+  `SlackIntegrationSettings { enabled: bool, default_channel:
+  Option<InterpString> }`.
+- Config layer: `lib/crates/fabro-config/src/layers/server.rs:237` —
+  `default_channel: Option<InterpString>`.
+- Resolve copy-through: `lib/crates/fabro-config/src/resolve/server.rs:365-369`
+  (clones the field into resolved server settings).
+- Startup resolution: `lib/crates/fabro-server/src/server.rs:2422-2430` —
+  `slack_settings.default_channel.as_ref().map(|value|
+  value.resolve(process_env_var)...)` feeding
+  `SlackService::new(bot, app, default_channel: Option<String>)`
+  (`server.rs:589-600`). The service posts interview prompts to it
+  (`server.rs:~657` `let Some(default_channel) = ... else return`, `:689`
+  `post_message`).
+- Demoted-field warning helper: `warn_if_demoted_template` in
+  `lib/crates/fabro-config/src/resolve/` (see its callers in
+  `resolve/cli.rs:50-76` for the exact usage pattern: field path string +
+  `Option<&str>` value).
+- Run-scope channels (untouched by this PR):
+  `run.notifications.<route>` slack channel and
+  `run.interviews.slack.channel`, both `InterpString`, variable-substituted at
+  run creation (`fabro-types/src/settings/run.rs`, `substitute_variables`).
+- Docs mentioning the field (all literal examples):
+  `docs/public/administration/server-configuration.mdx:453`,
+  `docs/public/human-tools/interviews.mdx:97`,
+  `docs/public/integrations/slack.mdx:112-117`.
+- OpenAPI: `docs/public/api-reference/fabro-api.yaml:13619-13623` models
+  `default_channel` as a nullable plain string in the relevant schema —
+  expected to need **no change**.
+
+## Implementation
+
+1. **Type change**: `fabro-types/src/settings/server.rs` and
+   `fabro-config/src/layers/server.rs` — `Option<InterpString>` →
+   `Option<String>`. Chase the compiler through the resolve copy-through and
+   any settings merge/serde helpers.
+2. **Demotion warning**: at the server-settings resolve site, call the
+   existing demoted-field warning helper with the field path
+   `server.integrations.slack.default_channel` and the literal value,
+   following the exact pattern of its existing callers. The warning must log
+   the field path and guidance only — never treat the value as sensitive
+   output beyond what the existing helper does.
+3. **Drop the startup resolve**: `fabro-server/src/server.rs` — pass the
+   literal through to `SlackService::new` directly; delete the
+   `resolve(process_env_var)` call and its error mapping. `SlackService`
+   itself is unchanged (it already takes `Option<String>`).
+4. **Verify (read-only) the interview routing preference**: confirm whether
+   the interview-prompt posting path prefers `run.interviews.slack.channel`
+   over the server default when both are set. If it does not, **do not build
+   routing changes** — record the finding in the PR description as a
+   follow-up observation.
+5. **Docs**: the three pages above already show literals; adjust wording only
+   if any implies interpolation (none is expected to). If the generated
+   options reference annotates the field's type, regenerate via
+   `cargo dev docs` and confirm `cargo dev docs check` is green.
+
+## Scope boundaries — deliberately NOT in this PR
+
+- **Run-scope notification and interview channel fields** — leave as-is; they
+  are the intended interpolating surface and are already correct.
+- **Slack credential resolution** (bot/app tokens via the vault at startup) —
+  leave as-is; unrelated to the channel field.
+- **Interview prompt routing behavior** — observe and report only (step 4);
+  changing which channel wins is separate product work.
+- **Any chat-integration plugin restructuring** — future work; this PR only
+  simplifies what that redesign will inherit.
+- **The config-field kind registry and its conformance tests** — separate
+  planned work; do not start it here.
+
+If work outside these boundaries seems genuinely required for this PR to
+compile or pass its tests, stop and state that in the PR description rather
+than expanding scope.
+
+## Tests (failing-first; hermetic — no ambient env dependence)
+
+- A literal channel (`#releases`) parses, merges, and reaches
+  `SlackService::new` unchanged.
+- A value containing a token-shaped span parses as a **literal** (no
+  resolution, no error) and emits the demoted-field warning naming
+  `server.integrations.slack.default_channel` (log-capture, matching how the
+  existing demotion-warning tests assert).
+- Startup wiring: with the field set, the service receives exactly the
+  configured string; with it absent, `None` (existing behavior preserved).
+- Existing Slack/server integration tests pass unchanged.
+- Serde shape: a settings round-trip of the field produces the same JSON/TOML
+  as before the type change (wire-invisibility).
+
+## Acceptance / verification
+
+- `cargo +nightly-2026-04-14 fmt --check --all`
+- `cargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings`
+- `cargo nextest run --workspace`
+- `cargo dev docs check`
+- No OpenAPI/wire change (assert: `docs/public/api-reference/fabro-api.yaml`
+  untouched; TypeScript client regeneration not needed).
+
+## Conventions
+
+- Plain-English commit messages, PR text, and comments — describe what the
+  change does; no internal planning identifiers or plan-file names in
+  anything that ships.
+- PR description must state plainly: the field only ever resolved env vars,
+  only at server startup, and was never documented as interpolable; anyone
+  with a token in the field gets a startup warning and the value is treated
+  as a literal; per-run channels (`run.notifications`, `run.interviews.slack`)
+  remain the interpolating surface; no wire change.
+
+
+## Completed stages
+- **toolchain**: succeeded
+  - 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`
+  - Output:
+    ```
+    cargo 1.96.0 (30a34c682 2026-05-25)
+    ```
+- **preflight_compile**: succeeded
+  - Script: `cargo check -q --workspace 2>&1`
+  - Output: (empty)
+- **preflight_lint**: succeeded
+  - Script: `cargo +nightly-2026-04-14 clippy -q --workspace --all-targets -- -D warnings 2>&1`
+  - Output: (empty)
+- **implement**: failed
+
+## Context
+- failure_class: deterministic
+- failure_signature: implement|deterministic|api_deterministic|openai|authentication
+
+
+# Simplify: Code Review and Cleanup
+
+Review all changes for reuse, quality, and efficiency. Fix any issues found. Feel free to use any sub agents you need.
+
+## Phase 1: Identify Changes
+
+Run git diff (or git diff HEAD if there are staged changes) to see what changed. If there are no git changes, review the most recently modified files that the user mentioned or that you edited earlier in this conversation. (You may already have the changes in context, if so, feel free to skip this part)
+
+## Phase 2: Launch Three Review Agents in Parallel
+
+Use the Agent tool to launch all three agents concurrently in a single message. Pass each agent the full diff so it has the complete context.
+
+### Agent 1: Code Reuse Review
+
+For each change:
+
+1. Search for existing utilities and helpers that could replace newly written code. Use Grep to find similar patterns elsewhere in the codebase — common locations are utility directories, shared modules, and files adjacent to the changed ones.
+2. Flag any new function that duplicates existing functionality. Suggest the existing function to use instead.
+3. Flag any inline logic that could use an existing utility — hand-rolled string manipulation, manual path handling, custom environment checks, ad-hoc type guards, and similar patterns are common candidates.
+
+Note: This is a greenfield app, so focus on maximizing simplicity and don't worry about changing things to achieve it.
+
+### Agent 2: Code Quality Review
+
+Review the same changes for hacky patterns:
+
+1. Redundant state: state that duplicates existing state, cached values that could be derived, observers/effects that could be direct calls
+2. Parameter sprawl: adding new parameters to a function instead of generalizing or restructuring existing ones
+3. Copy-paste with slight variation: near-duplicate code blocks that should be unified with a shared abstraction
+4. Leaky abstractions: exposing internal details that should be encapsulated, or breaking existing abstraction boundaries
+5. Stringly-typed code: using raw strings where constants, enums (string unions), or branded types already exist in the codebase
+
+Note: This is a greenfield app, so be aggressive in optimizing quality.
+
+### Agent 3: Efficiency Review
+
+Review the same changes for efficiency:
+
+1. Unnecessary work: redundant computations, repeated file reads, duplicate network/API calls, N+1 patterns
+2. Missed concurrency: independent operations run sequentially when they could run in parallel
+3. Hot-path bloat: new blocking work added to startup or per-request/per-render hot paths
+4. Unnecessary existence checks: pre-checking file/resource existence before operating (TOCTOU anti-pattern) — operate directly and handle the error
+5. Memory: unbounded data structures, missing cleanup, event listener leaks
+6. Overly broad operations: reading entire files when only a portion is needed, loading all items when filtering for one
+
+## Phase 3: Fix Issues
+
+Wait for all three agents to complete. Aggregate their findings and fix each issue directly. If a finding is a false positive or not worth addressing, note it and move on — do not argue with the finding, just skip it.
+
+When done, briefly summarize what was fixed (or confirm the code was already clean).
diff --git a/stages/006-simplify_fable@1/provider_used.json b/stages/006-simplify_fable@1/provider_used.json
new file mode 100644
index 000000000..a1f39be2f
--- /dev/null
+++ b/stages/006-simplify_fable@1/provider_used.json
@@ -0,0 +1,6 @@
+{
+  "mode": "agent",
+  "provider": "anthropic",
+  "model": "claude-fable-5",
+  "reasoning_effort": "xhigh"
+}
\ No newline at end of file
diff --git a/stages/006-simplify_fable@1/response.md b/stages/006-simplify_fable@1/response.md
new file mode 100644
index 000000000..b3d0a0dbf
--- /dev/null
+++ b/stages/006-simplify_fable@1/response.md
@@ -0,0 +1,24 @@
+## Summary
+
+The prior "implement" stage had failed on an API authentication error with **zero changes committed**, so I implemented the plan from scratch, then ran the three-agent simplify review and applied its findings.
+
+### Implementation
+- **Type demotion**: `default_channel: Option<InterpString>` → `Option<String>` in `fabro-types/src/settings/server.rs:261` and `fabro-config/src/layers/server.rs:237` (unused `InterpString` import removed from fabro-types).
+- **Demotion warning**: `resolve_integrations` in `fabro-config/src/resolve/server.rs` now calls `warn_if_demoted_template("server.integrations.slack.default_channel", ...)`, mirroring the adjacent GitHub arm.
+- **Dropped startup resolution**: `fabro-server/src/server.rs:2425` passes the literal straight to `SlackService::new`; the `resolve(process_env_var)` call and its error mapping are gone. Deleted the now-single-caller `display_interp` helper in `server/handler/system.rs` — the `/system/integrations` endpoint no longer does per-request template resolution.
+- **Tests**: literal pass-through + JSON wire-shape lock, token-stays-literal (mirroring `resolve_storage_root_keeps_template_token_literal`), and startup wiring tests (set → verbatim `#releases`; absent → `None`).
+
+### Step-4 finding (read-only, for the PR description)
+The interview-prompt posting path (`SlackService::handle_event`, server.rs:657) posts **only** to the server `default_channel` and never consults `run.interviews.slack.channel` — that run-scope field is parsed and resolved but has no consumer in the posting path. Follow-up product work, not changed here. Also worth noting in the PR: previously an unresolvable env token **failed startup**; now it flows through verbatim with a resolve-time warning — the intended "warn, don't break" behavior.
+
+### Simplify review results
+- **Fixed**: extracted `slack_test_vault_tokens()` to replace three verbatim copies of the vault-token fixture; aligned `system.rs` to the sibling `as_ref()` + clone-at-insert pattern.
+- **Skipped (with reason)**: dropping the JSON wire-shape assert (plan explicitly requires the serde wire-invisibility test) and folding the two wiring tests into one (plan requires both the set and absent cases).
+- Reviewers confirmed clean: no dead code, correct helper reuse, field-path string style matches siblings, private-field asserts match module test style, and the diff is a net perf win (startup and per-request resolution both removed).
+
+### Verification
+- `cargo +nightly-2026-04-14 fmt --check --all` — clean
+- `cargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings` — clean
+- `cargo nextest run --workspace` — 6961 passed (before final test-file cleanup; fabro-server/fabro-config re-run after: 984 passed)
+- `cargo dev docs check` — up to date; the three doc pages already show literals only
+- `docs/public/api-reference/fabro-api.yaml` — untouched; no wire change, no TS client regen needed
\ No newline at end of file