Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add gosha70/code-copilot-team --skill review-loopgit clone --depth 1 https://github.com/gosha70/code-copilot-teamWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gosha70/code-copilot-team/review-loop)<a href="https://agentmods.dev/skills/gosha70/code-copilot-team/review-loop"><img src="https://agentmods.dev/badge/skills/gosha70/code-copilot-team/review-loop/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gosha70/code-copilot-team/review-loop"><img src="https://agentmods.dev/badge/skills/gosha70/code-copilot-team/review-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00029 | $0.01579 |
| Opus 5.5 | $0.00012 | $0.00632 |
| Sonnet 5.5 | $0.00006 | $0.00316 |
| Haiku 4.5 | $0.00003 | $0.00158 |
Grade A, and why
review-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review Loop Protocol
Agent-driven peer review loop for Build and Plan phases. The Build agent drives the loop; the reviewer is a separate, read-only LLM.
Loop Architecture
The review loop is agent-driven: the Build agent calls /review-submit, which invokes review-round-runner.sh as a synchronous subprocess. The runner executes one round and returns. The agent reads the result and decides the next action. There is no long-running orchestrator.
Driver-as-submitter mode: under the auto-build loop (auto-build-loop.md), scripts/auto-build-loop.sh plays the /review-submit role — it initializes .cct/review/state.json, invokes the runner with CCT_REVIEW_BASE_REF=<phase-base> so the whole multi-commit phase diff is reviewed (the default remains HEAD~1), commits review fixes itself, and injects commit_ref into fixed dispositions. The runner's contract (exit codes, file formats, breakers, sandbox) is unchanged.
Agent completes work → /review-submit → review-round-runner.sh (one round)
↓
EXIT 0 (PASS) → loop-summary.json → /phase-complete
EXIT 1 (FAIL) → findings-round-N.json → agent fixes → /review-submit
EXIT 2 (BREAKER) → breaker-tripped.json → agent stops → /review-decide
File Contract
All review-loop artifacts live under .cct/review/ during the active loop.
| File | Written By | Purpose |
|---|---|---|
state.json |
/review-submit (init), runner (update) |
Loop state: round, attempt, findings, breaker counters |
findings-round-N.json |
Runner | Structured findings for round N |
resolution-round-N.json |
Build agent | Builder's response to each finding |
breaker-tripped.json |
Runner | Breaker context, awaiting /review-decide |
decision.json |
/review-decide |
Human's decision: approve, reject, or retry |
loop-summary.json |
Runner (PASS) or agent (bypass) | Final record of all rounds |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 136 lines · 29 tokens per session scan A 31b911c28d67
review-loop is a skill published in the GitHub repository gosha70/code-copilot-team (7 stars, last pushed 17d ago), licensed MIT. It adds 29 tokens to every session and 1,579 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
x-spec
A system-planning guide that turns a vague idea into a set of linked design documents. It defines goals, modules, interfaces, data, workflows, and ways to check the result.
x-audit-arch
A project-wide architecture review skill that checks whether code is placed in the right modules and whether important definitions have one reliable source.
x-multi-llm-align
A review process in which two sub-agents examine an API, data format, event schema, or workflow from their separate implementation perspectives. The user passes documents and feedback between them over multiple rounds.
pipeline-efficiency-benchmark
A repeatable benchmark for comparing different versions of a development pipeline. It runs the same public tasks in isolated workspaces and records scores, tokens, time, tool calls, and cost.
pipeline-eval-report
A project-level review and reporting process for checking what an agent or model produced during a pipeline run. It records quality scores, evidence, usage data such as tokens and time, and the stage where a failure began.
x-bug2rag
A knowledge-capture tool that turns reusable bug explanations into a local RAG collection, meaning a searchable store of text that an agent can retrieve later. It records the trigger, incorrect implementation, correct implementation, and observable difference.