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 daemonicai/teambuilder --skill teambuilder-review-gategit clone --depth 1 https://github.com/daemonicai/teambuilderWrote 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/daemonicai/teambuilder/teambuilder-review-gate)<a href="https://agentmods.dev/skills/daemonicai/teambuilder/teambuilder-review-gate"><img src="https://agentmods.dev/badge/skills/daemonicai/teambuilder/teambuilder-review-gate/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/daemonicai/teambuilder/teambuilder-review-gate"><img src="https://agentmods.dev/badge/skills/daemonicai/teambuilder/teambuilder-review-gate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00059 | $0.01099 |
| Opus 5 | $0.00030 | $0.00549 |
| Sonnet 5 | $0.00012 | $0.00220 |
| Haiku 4.5 | $0.00006 | $0.00110 |
Grade A, and why
teambuilder-review-gate 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 9d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run the Teambuilder Reviewer persona over a completed OpenSpec change and surface findings to the user before archival proceeds. Findings are non-blocking — the user decides whether to pause and address them.
Input: The active change name. The caller (typically /opsx:archive) invokes this skill before any archival steps run.
Steps
-
Check the Reviewer persona exists
Check whether
.claude/agents/reviewer.mdexists.If it does not exist:
- Stop immediately. Do not proceed with archival.
- Tell the user: "A Reviewer persona is required before archiving. Run
/teambuild:reviewerto create one, then re-run/opsx:archive." - Return control to the caller with a halt signal.
-
Collect the code diff
Goal: produce the diff of code changes made while working on this change.
a. Detect the base ref. Try these in order, and stop at the first that resolves:
git symbolic-ref --short refs/remotes/origin/HEAD(returns e.g.origin/main)origin/main, thenorigin/master(verify withgit rev-parse --verify <ref>)main, thenmaster(verify withgit rev-parse --verify <ref>)
If none resolve, stop and ask the user: "I couldn't detect a base branch automatically. What should I diff against? (e.g.,
origin/develop)"b. Compute the merge base:
git merge-base HEAD <base-ref>.c. Produce the diff:
git diff <merge-base>..HEAD.Also run
git diff HEADandgit diff --cachedto capture any uncommitted or staged changes; append them (labelled) if non-empty.d. Handle the "on the base branch" case. If
<merge-base>equalsHEAD(the user is working directly on the base branch), the committed diff is empty. Surface this to the user: "This change appears to be committed directly on the base branch — there is no feature-branch diff. Do you want me to diff the last N commits instead, or abort the review?" Wait for guidance before invoking the Reviewer. -
Invoke the Reviewer via the Agent tool
Use
subagent_type: "reviewer"(the persona file at.claude/agents/reviewer.mdis registered as a subagent by itsname:frontmatter — Claude Code loads it as the system prompt automatically, no need to tell the subagent to read it). If the Agent tool rejects"reviewer"as an unknown subagent type in this environment, fall back tosubagent_type: "general-purpose"and prepend "Read and adopt the persona defined in.claude/agents/reviewer.mdbefore starting." to the prompt.Prompt:
You are performing a per-change review at archive time for the change: <change-name> Read these change artifacts: - Proposal: <absolute path to openspec/changes/<name>/proposal.md> - Design: <absolute path to openspec/changes/<name>/design.md> - Specs: <absolute path to openspec/changes/<name>/specs/> - Tasks: <absolute path to openspec/changes/<name>/tasks.md> Code diff: <include the full git diff output here> Return your findings in two buckets: **Blocking** and **Warning**, using the review standards from your persona file. If you have no findings, return exactly: "No findings."Wait for the Reviewer subagent to return.
-
Handle the Reviewer's response
If the Reviewer returns "No findings.":
- Show "Reviewer: No findings." to the user.
- Return control to the caller to continue archival automatically.
If the Reviewer returns findings:
- Display all findings to the user clearly.
- Use the AskUserQuestion tool to ask: "Would you like to proceed with archiving, or pause to address these findings first?" with options:
Proceed with archiving,Pause — I'll address findings first. - If the user chooses Pause: stop here. Tell the user to re-run
/opsx:archiveafter addressing findings. Return a halt signal to the caller. - If the user chooses Proceed: return control to the caller to continue archival.
Guardrails
- Findings are non-blocking by design — the user, not the Reviewer, decides whether to address them before archiving.
- Halt archival if
.claude/agents/reviewer.mdis missing; do not fall back to a generic review. - Do not embed Reviewer findings into the archive itself — they are surfaced to the user at review time only.
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.
- 9d ago First seen · 80 lines · 59 tokens per session scan A 14bd66b7333a
teambuilder-review-gate is a skill published in the GitHub repository daemonicai/teambuilder (11 stars, last pushed 4mo ago), licensed MIT. It adds 59 tokens to every session and 1,099 once invoked, about $0.0003 per session on Opus 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 2026-08-30.
Other skills, from other repositories
migration-review
Review database migration files when a change adds or modifies paths under migrations/. Use it before merge to collect forward, rollback, locking, and data-safety evidence.
learn-from-sage
Detection-gap (miss) analysis for Code Review Sage. Learn from shipped fixes, acted-on human comments, and design outcomes to close reviewer blind spots. Inline during review stages a candidate; a human triggers a one-shot AI consolidation into the live ruleset.
github-review-pr
Review GitHub pull requests with detailed, multi-perspective code analysis using parallel subagents. Use this skill whenever the user wants to review a PR, asks for code review on a pull request, mentions "review PR", "check this PR", "look at pull request", or references a PR number or GitHub PR URL. Do NOT use for…
lean-review
CocoLean diff-scoped over-engineering audit — scans uncommitted git diff and applies five classification tags (delete/stdlib/native/yagni/shrink) to identify unnecessary surface area before commit.
review-export
Export CocoReview reports as markdown, html, or pdf-ready status artifacts. Usage: $review export markdown|html|pdf [report-path].
no-mistakes
Validate committed feature-branch changes through the no-mistakes pipeline: intent, rebase, review, test, docs, lint, push, PR, and CI. Use when the user asks to run no-mistakes, ship safely, validate before pushing, or gate a change before it reaches upstream.