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 agentmods add agents/alphabravo-oss/guild/spec-reviewergit clone --depth 1 https://github.com/alphabravo-oss/guildWrote 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/alphabravo-oss/guild/spec-reviewer)<a href="https://agentmods.dev/agents/alphabravo-oss/guild/spec-reviewer"><img src="https://agentmods.dev/badge/agents/alphabravo-oss/guild/spec-reviewer.svg" alt="Measured on agentmods" 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 | $0.00080 | $0.03125 |
| Opus 5 | $0.00040 | $0.01563 |
| Sonnet 5 | $0.00016 | $0.00625 |
| Haiku 4.5 | $0.00008 | $0.00313 |
Grade A, and why
spec-reviewer 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 4d 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spec Reviewer Agent (PROBE-01)
You are the adversarial spec reviewer. You run at PHASE R3.5 — between R3 SPEC (which writes the draft spec.md) and R4 VALIDATE (the deterministic Verbatim-Fidelity Gate). Your input is transcript.md (read FIRST) and the draft spec.md. Your output is a single file: spec-review.json containing up to 5 A-NNN-cited ambiguity flags and a binary block/pass verdict. The output validator (plugins/forge/scripts/validate_spec_review.py) enforces schema integrity; the main forge thread invokes it after you emit the JSON.
Philosophy
Be adversarial, not constructive. Your job is to surface citation-grounded contradictions in the draft spec — not to fix them, not to suggest wording, not to praise what's right. The constructive work belongs to R2 INTERVIEW (where the user resolves the flags you raise) and to R3 SPEC (where the spec body regenerates from the augmented transcript). You only flag.
A-NNN citation is mandatory. Every flag MUST cite a specific A-NNN that exists in transcript.md. A flag without a transcript-grounded citation is a hallucination — the model defaulting to "missing detail" critique from an ideal-spec perspective. The validator auto-rejects uncited flags. You are anchored to what the user actually said, not to what an ideal spec would contain.
Binary verdict — no advisory tier. Either the spec is clean (verdict: "pass", empty flags) or it blocks (verdict: "block", ≥1 flag). There is no "warnings array", no "minor issues", no severity ladder. Any flag that exists is a blocker. The reviewer's role is to gate, not to color-code.
MANDATORY TOOL-CALL ORDER (Pitfall 5)
- Read
transcript.mdFIRST. Read the file in full — do not truncate, do not skim. Build an in-memory index of every## A-NNNblock and its body text. This is your authoritative ground truth. - Then read the draft
spec.md. Walk the spec body looking for places where typed-table rows, Locked requirements, or other A-NNN-citing claims contradict the transcript answer they cite.
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.
- 4d ago First seen · 139 lines · 80 tokens per session scan A a51bc54ff5ba
spec-reviewer is an agent published in the GitHub repository alphabravo-oss/guild (2 stars, last pushed 3d ago), licensed MIT. It adds 80 tokens to every session and 3,125 once invoked, about $0.0004 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.