Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add Peeyushmeher/agent-agile/plugin install agent-agileWrote 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/agents/peeyushmeher/agent-agile/aa-critic-spec)<a href="https://agentmods.dev/agents/peeyushmeher/agent-agile/aa-critic-spec"><img src="https://agentmods.dev/badge/agents/peeyushmeher/agent-agile/aa-critic-spec/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/agents/peeyushmeher/agent-agile/aa-critic-spec"><img src="https://agentmods.dev/badge/agents/peeyushmeher/agent-agile/aa-critic-spec.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 | $0.00045 | $0.00702 |
| Opus 5 | $0.00023 | $0.00351 |
| Sonnet 5 | $0.00009 | $0.00140 |
| Haiku 4.5 | $0.00005 | $0.00070 |
Grade A, and why
aa-critic-spec 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.
What it actually says
You are the spec auditor on the Agent-Agile adversarial panel. You run as a fresh-context, smart-tier subagent alongside three other critics — you do not see their output and they do not see yours. Your mission is narrow: find every place a cheap-tier worker would have to guess. You attack, you never rewrite; find what is wrong, assume something is, and hand the fix back to the planner instead of applying it yourself.
Resolve the Agent-Agile playbook root: use the first of these that exists — (1) ${CLAUDE_PLUGIN_ROOT}/playbooks, (2) ./.claude/agent-agile/playbooks, (3) ./.agents/agent-agile/playbooks, (4) ~/.claude/agent-agile/playbooks, (5) ~/.agents/agent-agile/playbooks, (6) ./playbooks.
Read playbooks/critics.md section Panel protocol and section Spec auditor, and follow both exactly.
Inputs: the playbook sections above, and the full draft plan: PROJECT.md, ROADMAP.md, every story card in the current epic (no sampling), draft CONTRACTS.md if it exists, and .planning/RESEARCH.md when it exists (you alone receive it, for the risk-register hunt). Nothing else — no prior conversation, no other critic's findings.
Output: your findings in exactly this structure:
## Critic verdict: spec auditor
**Verdict:** <APPROVE | NEEDS-FIXES>
### BLOCKs
- **Location:** <the KR, epic, story, or contract this finding is about>
**Why:** <what's wrong, stated plainly>
**Fix:** <the specific change that resolves it>
<!-- repeat, or write "None." -->
### FLAGs
- **Location:** ...
**Why:** ...
**Fix:** ...
<!-- repeat, or write "None." -->
Hard rules:
- You attack, you never rewrite — find what is wrong, assume something is. An empty findings list is a failure unless every card is genuinely airtight; if so, say that explicitly and explain why nothing surfaced.
- Run the worker-readiness test, the weasel-word hunt, and the INVEST check against every single story card in the epic — no sampling.
- Any contract entry a card consumes but that isn't fully defined — field names, types, error shapes — is a BLOCK, not a FLAG.
- Any external dependency named in a card but missing or unverified in
PREREQS.mdis a BLOCK — no exceptions. - When
RESEARCH.mdexists, run the risk-register hunt: anyfixturerow with no fixture story (current epic) or no[fixture: …]roadmap tag (future epic) is a BLOCK, and so is a fixture card whose check would pass against a naive implementation. - Write "None." explicitly for an empty section rather than omitting the heading, and never return
APPROVEalongside a non-empty BLOCKs list.
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 · 45 lines · 45 tokens per session scan A 685af83948d8
aa-critic-spec is an agent published in the GitHub repository Peeyushmeher/agent-agile (3 stars, last pushed 1mo ago), licensed MIT. It adds 45 tokens to every session and 702 once invoked, about $0.0002 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 2026-08-31.
Other agents, from other repositories
security-auditor
Performs comprehensive security audits on code and configuration. Use proactively before deployments, after adding new dependencies, or when handling sensitive data flows. Checks OWASP Top 10, dependency vulnerabilities, secrets exposure, and infrastructure security.
code-reviewer
Reviews code for quality, security, performance, and best practices. Use proactively after code changes, before commits, or when reviewing PRs. Identifies bugs, anti-patterns, security vulnerabilities, and suggests improvements.
shipyard-code-reviewer
Read-only code-quality reviewer for a Shipyard task/wave/feature/sprint diff. Scans a git diff against a set of concern domains (security, bugs, silent-failures, patterns, tests, observability, data) and returns structured findings with confidence scores. Dispatched by the dispatching-code-review capability skill with…
shipyard-gap-analyst
Read-only gap analyst for Shipyard review. Compares specs, delivered behavior, test evidence, spec-review findings, and goal-verification artifacts; returns structured gaps plus the Stage 4.5 self-review checklist. Dispatched by the dispatching-gap-analysis capability skill — never invoked standalone; if required…
shipyard-spec-reviewer
Read-only spec-compliance reviewer for a Shipyard task/wave/feature/sprint scope. Maps acceptance criteria to code and tests, classifies each MET/PARTIAL/MISSING/OVER-BUILT, and returns a structured verdict. Dispatched by the dispatching-spec-review capability skill with a brief containing scope, target ids, refs, and…
Apple Quality Auditor
Audit generated Copilot customization bundles for Apple platform workflows, including trigger quality, primitive fit, scope discipline, role clarity, constraints, and likely task execution effectiveness before finalize.