feat(model-routing): wire blocking model gate into 12 reflection orchestrators

This commit is contained in:
Bastien Chanot
2026-07-15 11:11:29 +02:00
parent f2dd361bd5
commit 06413d9eb5
12 changed files with 75 additions and 0 deletions
+7
View File
@@ -22,6 +22,13 @@ allowed-tools:
# /audit-delta — Incremental multi-axis code audit # /audit-delta — Incremental multi-axis code audit
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
Audit only what changed since the last run, on the axes the user picks. Audit only what changed since the last run, on the axes the user picks.
Per axis: **audit → approval gate → fix → re-verify → marker update**, Per axis: **audit → approval gate → fix → re-verify → marker update**,
strictly in that order, one axis fully closed before the next starts. strictly in that order, one axis fully closed before the next starts.
+4
View File
@@ -20,6 +20,10 @@ allowed-tools:
- Agent - Agent
--- ---
MODEL GATE (blocking): run `$HOME/.claude/lib/model-gate.md` BEFORE loading
the agent below. Verdict `small` → STOP — print the gate's remedy, end the
turn, do not load the agent.
Load and follow strictly: Load and follow strictly:
- $HOME/.claude/agents/bugfixer.md - $HOME/.claude/agents/bugfixer.md
+4
View File
@@ -20,6 +20,10 @@ allowed-tools:
- AskUserQuestion - AskUserQuestion
--- ---
MODEL GATE (blocking): run `$HOME/.claude/lib/model-gate.md` BEFORE loading
the agent below. Verdict `small` → STOP — print the gate's remedy, end the
turn, do not load the agent.
Load and follow strictly: Load and follow strictly:
- $HOME/.claude/agents/code-cleaner.md - $HOME/.claude/agents/code-cleaner.md
+4
View File
@@ -20,6 +20,10 @@ allowed-tools:
- Agent - Agent
--- ---
MODEL GATE (blocking): run `$HOME/.claude/lib/model-gate.md` BEFORE loading
the agent below. Verdict `small` → STOP — print the gate's remedy, end the
turn, do not load the agent.
Load and follow strictly: Load and follow strictly:
- $HOME/.claude/agents/feater.md - $HOME/.claude/agents/feater.md
+7
View File
@@ -22,6 +22,13 @@ allowed-tools:
# /geo — GEO (AI-search) audit + fix dispatcher # /geo — GEO (AI-search) audit + fix dispatcher
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
Dispatches the `geo-analyzer` subagent (audit + fix bundle), then applies Dispatches the `geo-analyzer` subagent (audit + fix bundle), then applies
the bundle from THIS main loop at **L1** — same shape as `/web-validate` the bundle from THIS main loop at **L1** — same shape as `/web-validate`
and `/seo`. The analyzer never edits files: it emits a `## FIX BUNDLE` and `/seo`. The analyzer never edits files: it emits a `## FIX BUNDLE`
+7
View File
@@ -22,6 +22,13 @@ allowed-tools:
# /harden — web hardening audit # /harden — web hardening audit
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
This skill orchestrates a narrow-scope hardening audit: TLS + security This skill orchestrates a narrow-scope hardening audit: TLS + security
headers + redirects + canonical + custom 404 + server configs. It headers + redirects + canonical + custom 404 + server configs. It
reuses the `seo-analyzer` agent with a **strict scope filter** to avoid reuses the `seo-analyzer` agent with a **strict scope filter** to avoid
+7
View File
@@ -7,6 +7,13 @@ allowed-tools: Read, Write, Edit, Bash, Grep, Glob
# ORCHESTRATOR: INIT PROJECT # ORCHESTRATOR: INIT PROJECT
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
## REQUEST ## REQUEST
$ARGUMENTS $ARGUMENTS
+7
View File
@@ -7,6 +7,13 @@ allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Agent, Skill
# ORCHESTRATOR: ONBOARD # ORCHESTRATOR: ONBOARD
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
## REQUEST ## REQUEST
$ARGUMENTS $ARGUMENTS
+7
View File
@@ -23,6 +23,13 @@ allowed-tools:
# /seo — parallel SEO + GEO dispatcher # /seo — parallel SEO + GEO dispatcher
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
This skill orchestrates TWO specialist agents running in parallel, then This skill orchestrates TWO specialist agents running in parallel, then
merges their output into a single `.claude/audits/SEO.md` report. It is the main merges their output into a single `.claude/audits/SEO.md` report. It is the main
entry point for any SEO/GEO work on a web project. entry point for any SEO/GEO work on a web project.
+7
View File
@@ -7,6 +7,13 @@ allowed-tools: Read, Write, Edit, Bash, Grep, Glob
# ORCHESTRATOR: SHIP FEATURE # ORCHESTRATOR: SHIP FEATURE
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
## REQUEST ## REQUEST
$ARGUMENTS $ARGUMENTS
+7
View File
@@ -23,6 +23,13 @@ allowed-tools:
# /tour — grouped multi-axis sweep (clean + security + reconcile + doc) # /tour — grouped multi-axis sweep (clean + security + reconcile + doc)
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
One pipeline per project: **security → clean → re-verify → reconcile → One pipeline per project: **security → clean → re-verify → reconcile →
doc → convergence re-audit**, looping until a full pass applies zero new doc → convergence re-audit**, looping until a full pass applies zero new
fixes. Auto mode by design: fixes are committed on a dedicated fixes. Auto mode by design: fixes are committed on a dedicated
+7
View File
@@ -21,6 +21,13 @@ allowed-tools:
# /web-validate — web standards audit (W3C + WCAG) # /web-validate — web standards audit (W3C + WCAG)
## MODEL GATE (blocking — run before any other step)
Run `$HOME/.claude/lib/model-gate.md`. Reflection here (planning, audit
judgment, loop decisions) requires Fable/Opus. Verdict `small` → STOP: the
gate prints the remedy; end the turn — no later step, no dispatch. Nominal
(big) path is silent.
This skill orchestrates a narrow-scope standards audit : This skill orchestrates a narrow-scope standards audit :
- **W3C HTML validity** — validator.nu API (FULL) or `html-validate` / - **W3C HTML validity** — validator.nu API (FULL) or `html-validate` /