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 skills/tmj-90/gaffer/prdnpx skills add tmj-90/gaffer --skill prdgit clone --depth 1 https://github.com/tmj-90/gafferWhat 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 | $0.00070 | $0.00832 |
| Opus 5 | $0.00035 | $0.00416 |
| Sonnet 5 | $0.00014 | $0.00166 |
| Haiku 4.5 | $0.00007 | $0.00083 |
Grade A, and why
prd 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 2d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write evidence-gated product requirements
A PRD without a real problem, a specific user, and a measurable metric is a feature wish. Refuse to draft until those three are answered.
Forcing questions (walk one at a time — do not batch)
- Problem — What user problem does this solve, and how do you know it exists? (Support tickets, interview quotes, funnel data — "the CEO wants it" is not evidence.)
- User — Who specifically has this problem? (Segment, role, frequency of pain. "Everyone" is not a user.)
- Metric — What single number moves if this works, by how much, measured where?
- Alternatives — What do these users do today instead? Why is that not enough?
- Non-goals — What adjacent asks are explicitly out of scope for v1?
Drafting gate
Refuse to draft if Q1 (problem), Q2 (user), or Q3 (metric) is unknown, circular, or "we'll figure it out."
Instead: output the open questions and the cheapest way to answer each (5 customer interviews, a funnel query, a fake-door test, a 1-day prototype). A week of discovery is cheaper than shipping the wrong feature.
Required sections (every PRD must have all of these)
- Problem statement — with the evidence from Q1.
- Target user and segment — from Q2; who is explicitly NOT the target.
- Goals and explicit non-goals — from Q5.
- User stories with acceptance criteria — "As a [user], I want [action] so that [outcome]. Done when [testable condition]."
- Success metric + threshold + measurement source — from Q3; how and when you'll measure.
- Open questions — unresolved assumptions that need an answer before or during build.
- Out of scope — explicit list; protects from scope creep in delivery.
Acceptance criteria format
Given [context / precondition]
When [action]
Then [observable outcome]
Every user story has at least one AC. ACs are testable — a QA engineer can verify them without asking questions.
Steps
- Ask the forcing questions one at a time. Wait for answers. Don't proceed if answers to Q1–3 are vague.
- Apply the drafting gate. If blocked, output the open questions + cheapest resolution path.
- Draft the PRD in the required-sections format.
- Emit the completion checklist at the end — mark each section done or flag what's missing.
- Review with stakeholders before handing to engineering — especially the success metric.
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.
- 2d ago First seen · 68 lines · 70 tokens per session scan A a428c79f89ad
prd is a skill published in the GitHub repository tmj-90/gaffer (2 stars, last pushed 8d ago), licensed Apache-2.0. It adds 70 tokens to every session and 832 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 2026-08-31.
Other skills, from other repositories
mobile-developer
Expert knowledge in mobile app development for iOS, Android, and cross-platform solutions.
skill-creator
Create or update AnkaLoop skills. Use when designing, structuring, or packaging skills with scripts, references, and assets. This skill should be used when users want to create a new skill (or update an existing skill) that extends AnkaLoop's capabilities with specialized knowledge, workflows, or tool integrations.
ai-engineer
Expert knowledge in AI/ML development, model deployment, and MLOps practices.
gh-ci-analyzer
使用 gh CLI 分析 GitHub Actions CI 日志，提供结构化的故障诊断和报告.
security-best-practices
Security best practices and vulnerability prevention guidelines.
react-developer
Expert knowledge in React development, hooks, state management, and modern React patterns.