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.
git clone --depth 1 https://github.com/tercel/spec-forgeWrote 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/commands/tercel/spec-forge/review)<a href="https://agentmods.dev/commands/tercel/spec-forge/review"><img src="https://agentmods.dev/badge/commands/tercel/spec-forge/review/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/commands/tercel/spec-forge/review"><img src="https://agentmods.dev/badge/commands/tercel/spec-forge/review.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 | $0.00022 | $0.00505 |
| Opus 5.5 | $0.00009 | $0.00202 |
| Sonnet 5.5 | $0.00004 | $0.00101 |
| Haiku 4.5 | $0.00002 | $0.00051 |
Grade A, and why
review 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior specification reviewer responsible for ensuring spec-forge generated documents meet quality standards.
Your task is to review the specifications for: $ARGUMENTS
Workflow
Step 1: Determine Target
Parse $ARGUMENTS:
- If a feature name is provided, use it to locate
docs/{feature_name}/tech-design.mdanddocs/features/*.md - If empty, scan
docs/for the most recent tech-design and all feature specs - Verify at least one spec document exists
Step 2: Launch Review
Launch Task(subagent_type="general-purpose") with the following prompt:
You are a senior specification reviewer. Your task is to review spec-forge generated documents for quality, completeness, and internal consistency, and optionally auto-fix issues found.
Target feature: {feature_name or "auto-detect"}
Read the review skill definition at:
skills/review/SKILL.md
Follow every step of the workflow exactly. Skip path resolution in Step 1 (already resolved above). Start by asking the user about review scope and auto-fix preference (Step 1 questions), then proceed through all remaining steps.
Key rules:
- Every finding must cite specific files and sections — no vague complaints
- Check consistency between tech-design and feature specs (API signatures, component boundaries, data models)
- Classify findings by severity: Critical, Major, Minor
- Auto-fix only modifies cited sections — never restructures entire documents
- When domain knowledge is missing, leave
<!-- REVIEW: {question} -->comments instead of guessing - Maximum 2 review-fix iterations
- Be honest — don't inflate findings and don't fabricate issues
Step 3: Present Results
After the sub-agent returns, display the summary and suggest next steps:
Next steps:
Fix remaining issues manually if any
Re-run /spec-forge:review {feature_name} after manual fixes to verify
/code-forge:plan @docs/features/{component-name}.md → Start implementation
/spec-forge:audit {feature_name} → Full audit including code alignment
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 · 57 lines · 22 tokens per session scan A 77c60fc8e90f
review is a command published in the GitHub repository tercel/spec-forge (10 stars, last pushed 1mo ago), licensed MIT. It adds 22 tokens to every session and 505 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 2026-08-31.
Other commands, from other repositories
cs-grill-agent-launcher
Matt Pocock docs-anchored grill for an agent-launcher goal — walks the phase's forcing questions ONE at a time, each with a recommended answer and a citation to a reference doc, refusing to advance on a fuzzy input. Use to pressure-test a CMA plan before launching.
promote
The promote loop — scan your transcripts for what you keep asking, pick a pattern, draft the skill, test its triggers, publish the release package.
rescue-an-account
Workflow recipe — diagnose an at-risk customer and build the full save play through to renewal by chaining 4 skills.
psql-query
Run ad-hoc PostgreSQL analytics queries against dev/test database.
prompt-optimizer
Analyze and rewrite a prompt to maximize clarity, specificity, and output quality.
checklist
Generate a custom checklist for the current feature based on user requirements.