Add headless Claude Code (Opus/Sonnet) as a delegation-lane Producer #37
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,17 @@ | ||
| --- | ||
| name: candidate-reviewer | ||
| description: Independent read-only reviewer for ONE frozen Candidate Artifact. Input is a checkoutPath, runId, protocolVersion, and the review brief (spec, success criteria, findings to re-check); output is a structured verdict. Reads the exact anchored bytes through reviewCandidate and never edits, decides, or integrates. | ||
| tools: Read, Grep, Glob, mcp__plugin_claude-architect_runtime__reviewCandidate | ||
| model: opus | ||
| --- | ||
|
|
||
| You review exactly one frozen candidate. You share no context with the Producer that made it: your only inputs are the fields in your prompt and the runtime's own evidence. Ignore repository lore, CLAUDE.md content, and git status injected into your context. | ||
|
|
||
| Your prompt provides: `checkoutPath`, `runId`, `protocolVersion`, the Delegation Spec's objective, success criteria, and `review.focus`, and — on a re-review — the numbered findings list from the previous round. | ||
|
|
||
| 1. Call `reviewCandidate` with `checkoutPath`, `runId`, and `protocolVersion` exactly as given. Read the unredacted patch, the changed-path manifest, and the verification evidence it returns. That is the entire candidate; the Producer's summary is a correlation aid, never evidence. | ||
| 2. Use `Read`/`Grep`/`Glob` only to understand code the patch touches or depends on. Never modify anything. | ||
| 3. Give two verdicts, each with the evidence that decides it: **spec compliance** (every success criterion met, scope honored, nothing outside the allowlist) and **quality** (Critical / Important / Minor findings with file and line). On a re-review, mark each prior finding ADDRESSED or NOT ADDRESSED, then list new breakage in this candidate only. | ||
| 4. End with a single line: `RECOMMEND accept` or `RECOMMEND revision-requested`. It is a recommendation: only the architect calls `decideCandidate`, and only the configured decision authority records the decision. | ||
|
|
||
| Never call `decideCandidate` or `integrateCandidate`, never re-run the Producer, never propose patching the candidate yourself. |
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -27,7 +27,7 @@ The plugin is designed for macOS, Linux, and Windows process/runtime operation. | |
|
|
||
| ## Network destinations | ||
|
|
||
| There is no plugin-maintained fixed destination list. A cloud Producer CLI contacts the provider configured by that CLI: Codex normally uses its configured OpenAI service; OpenCode, Pi, Pythinker, Antigravity CLI, and headless Claude Code can use various cloud or local endpoints. Claude Code separately contacts its configured Anthropic/model service. Verification commands may contact destinations only when their spec allows network, subject to effective platform enforcement. Codex's coding sandbox is configured with network disabled. Provider authentication, telemetry, transport, and retention are governed by the selected CLI/provider. | ||
|
There was a problem hiding this comment. 🔒 Security & Privacy | 🟡 Minor | ⚡ Quick win 🔎 Supported by static analysis🏁 Script executed: #!/bin/bash
set -euo pipefail
rg -n -C 5 \
'CLAUDE_REQUIRED_ENV|ANTHROPIC_BASE_URL|endpoint|local provider|provider' \
src/producers/claude-adapter.ts \
tests/runtime/claude-adapter.test.ts \
|| trueRepository: PyModel/claude-architect Length of output: 3915 🏁 Script executed: #!/bin/bash
set -euo pipefail
printf '%s\n' '--- applicable repository conventions ---'
for f in /tmp/coderabbit-repo-knowledge/pymodel-claude-architect-d6e41926/*/*.md; do
[ -f "$f" ] || continue
case "$f" in
*/*.md) printf '\n### %s\n' "$f"; head -80 "$f";;
esac
done
printf '%s\n' '--- Claude adapter ---'
cat -n src/producers/claude-adapter.ts | sed -n '1,225p'
printf '%s\n' '--- Claude tests: configuration and invocation assertions ---'
cat -n tests/runtime/claude-adapter.test.ts | sed -n '1,180p'
cat -n tests/runtime/claude-adapter.test.ts | sed -n '330,430p'
printf '%s\n' '--- repository-owned Claude endpoint/config references ---'
rg -n -C 3 \
'Claude Code|claude-adapter|ANTHROPIC_BASE_URL|ANTHROPIC_API_KEY|CLAUDE_CONFIG_DIR|local provider|endpoint|model service' \
docs src tests README.md package.json 2>/dev/null || trueRepository: PyModel/claude-architect Length of output: 50383 🏁 Script executed: #!/bin/bash
set -euo pipefail
printf '%s\n' '--- environment policy contract ---'
ast-grep outline src/runtime/environment-policy.ts
cat -n src/runtime/environment-policy.ts | sed -n '1,240p'
printf '%s\n' '--- invocation environment construction and Producer contract ---'
rg -n -C 8 \
'requiredEnv|buildEnvironment|environmentDependencies|ANTHROPIC_BASE_URL|CLAUDE_CONFIG_DIR|ANTHROPIC_API_KEY' \
src/runtime src/producers tests/runtime/claude-adapter.test.ts
printf '%s\n' '--- privacy statement ---'
cat -n docs/PRIVACY.md | sed -n '1,45p'Repository: PyModel/claude-architect Length of output: 50380 🌐 Web query:
💡 Result: To configure Claude Code to use a local or custom endpoint, you must set the Citations:
Remove the local-provider claim for headless Claude Code. The delegated 📍 Affects 2 files
🤖 Prompt for AI AgentsSource: Path instructions |
||
|
|
||
| ## Persistent state locations | ||
|
|
||
|
|
||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
Document Claude's supported overrides separately.
ClaudeAdapter.buildInvocationconsumesproducerOverrides.modelandproducerOverrides.reasoningEffort. The supplied contract does not showthinkingorvariantsupport for Claude. Rewrite this sentence with per-harness fields so users do not send unsupported Claude overrides.As per path instructions, Markdown prose must agree with executable contracts.
🤖 Prompt for AI Agents
Source: Path instructions