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/nicograef/handbook/write-prdnpx skills add nicograef/handbook --skill write-prdgit clone --depth 1 https://github.com/nicograef/handbookWrote 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/nicograef/handbook/write-prd)<a href="https://agentmods.dev/skills/nicograef/handbook/write-prd"><img src="https://agentmods.dev/badge/skills/nicograef/handbook/write-prd.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 | $0.00051 | $0.01248 |
| Opus 5 | $0.00026 | $0.00624 |
| Sonnet 5 | $0.00010 | $0.00250 |
| Haiku 4.5 | $0.00005 | $0.00125 |
Grade A, and why
write-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 today.
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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write a PRD
- Skip a step only when its output already exists — e.g. clarification done, codebase understood.
- Never skip exploration.
- Named prose exception: the PRD's Problem Statement and User Stories stay prose.
- Reason: PRD readers are non-technical. Do not "fix" those two slots into bullets.
- The ≤ 20-word sentence cap still applies to them.
- Everything else follows the output style contract.
Workflow
1. Gather the problem description
- Ask the user for a long, detailed description of the problem to solve.
- Ask for any potential ideas for solutions.
- Scope triage: stop before clarification rounds if the request bundles multiple independent subsystems or features.
- Independent means each one could ship and be useful on its own.
- Help the user split such a request into separate PRDs.
- Each split gets its own PRD and, later, its own plan.
- Continue this workflow only once the scope is a single, cohesive PRD.
2. Explore the codebase
- Explore the repo to verify the user's assertions.
- Understand the current state of the codebase.
3. Clarify ambiguities
- Resolve unknowns through 0–3 rounds of structured questions.
- Follow the canonical clarification question rules.
4. Propose approaches
- Work out 2–3 candidate solution approaches once ambiguities are resolved.
- Give each approach its trade-offs: effort, risk, reversibility, fit with the existing codebase.
- Then run the ask gate over them; the survivor becomes the recorded Implementation Decisions entry, with its reasoning.
5. Design modules
- Sketch the major modules you will build or modify to complete the implementation.
- Actively look for deep modules you can extract and test in isolation.
- A deep module hides substantial functionality behind a small, stable interface.
- Inject dependencies instead of constructing them.
- Return values instead of mutating inputs.
- Keep the interface surface small.
- Present the module design as part of the flow; do not block on a separate confirmation.
- Decide the test scope yourself, based on the design.
- Raise test scope as a clarification question only if it is a genuine judgment call the codebase cannot answer.
- Keep any such question inside the rounds from step 3.
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.
- today Changed · -7 lines 6f5934cafff2
- 4d ago First seen · 154 lines · 51 tokens per session scan A 60f96ad03383
write-prd is a skill published in the GitHub repository nicograef/handbook (2 stars, last pushed yesterday), licensed MIT. It adds 51 tokens to every session and 1,248 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
instrument-data-to-allotrope
Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…
exploratory-data-analysis
Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
read
Reads URLs and PDFs by fetching source content, defaulting to concise summaries for plain read requests and clean Markdown when asked to convert, save, quote, cite, or feed downstream work. Use when users ask in any language to read, fetch, check, summarize, quote, cite, convert, or save a URL or PDF. Not for local…
docx-comment-reply
Reply to comments (批注) in Word .docx/.doc files: extract comment context, draft replies, write threaded replies back, and validate OOXML.