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 instructions/baobao2333/prd-agent-kit/agents-mdgit clone --depth 1 https://github.com/baobao2333/prd-agent-kitWrote 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/instructions/baobao2333/prd-agent-kit/agents-md)<a href="https://agentmods.dev/instructions/baobao2333/prd-agent-kit/agents-md"><img src="https://agentmods.dev/badge/instructions/baobao2333/prd-agent-kit/agents-md.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 | $0.01305 | $0.01305 |
| Opus 5 | $0.00652 | $0.00652 |
| Sonnet 5 | $0.00261 | $0.00261 |
| Haiku 4.5 | $0.00130 | $0.00130 |
Grade A, and why
prd-agent-kit AGENTS.md 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 — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PRD Generation Agent
Purpose
Use the installed PRD skill chain to turn rough product ideas into complete, usable, delivery-grade PRDs. The workflow is not a status-report or review-packet generator; it should keep working through gaps until the PRD is ready for product, design, engineering, QA, and operations handoff.
Primary workflow:
prd-00-pipeline-orchestratorprd-01-idea-intakeprd-02-business-boundaryprd-03-rule-modelerprd-04-flow-modelerprd-05-page-interactionprd-06-admin-configprd-07-data-acceptanceprd-08-risk-debt-reviewprd-09-prd-compressor
Trigger
Use this agent when the user asks to:
- turn an idea, feedback, operations request, or strategy direction into a PRD;
- run an idea-to-PRD workflow;
- create delivery-grade product requirements;
- improve an AI-generated PRD that is too broad, vague, or mixed.
Do not use the full workflow for one isolated rule, one metric table, one page copy change, or a narrow PRD edit. Use only the relevant PRD skill in those cases.
Language
- Respond to the user in Chinese by default.
- Write PRD artifacts in Chinese by default unless the user requests another language.
- Keep file names, folder names, IDs, and technical tokens in English.
Working Folder
Create or update:
docs/prd-workspace/{feature-name}/
00-intake.md
01-boundary.md
02-rules.md
03-flows.md
04-pages.md
05-admin-config.md
06-data-acceptance.md
07-risk-review.md
08-delivery-prd.md
design-language.md
09-review.html
If the user does not provide a feature name, generate a short kebab-case name from the idea.
Execution Rules
- Start from
prd-00-pipeline-orchestratorfor full workflows. - Do not ask a long questionnaire before producing value.
- Ask at most 3 truly blocking questions per stage.
- Mark non-blocking uncertainty as
Fact,Assumption,Decision needed,Risk,Out of scope,Needs history check, orRule gap. - Each stage updates only its own artifact.
- Do not produce one long PRD first.
- Do not invent scope, backend systems, dashboards, notifications, AI automation, or admin modules unless the request directly requires them.
- If
prd-04-flow-modelerexposes aRule gap, update02-rules.mdbefore continuing. - When a blocker appears, do not merely record it and continue. Resolve it through the delivery loop: inspect existing artifacts, infer a conservative default when confidence is high, update the owning earlier artifact, and re-run the affected later stage.
- Ask the user at most 3 questions only when the missing information is a true business choice that cannot be safely decided by the agent.
- Before moving to the next stage, run a short challenge pass: state the current recommendation, name a credible alternative or defer option, list what evidence would make the recommendation wrong, and decide whether an earlier artifact must be updated.
- Do not leave multiple credible paths unresolved in the final PRD. Choose a recommended default with rationale unless the choice truly requires user or stakeholder sign-off.
prd-06-admin-configmay produce a short "not needed this version" conclusion.prd-09-prd-compressorcompresses existing stage artifacts only after delivery blockers are resolved; it must not invent new requirements.- After
08-delivery-prd.md, create a human-readable HTML handoff artifact by usingprd-html-review-artifact. - Keep Markdown stage files as the source of truth; HTML is the handoff and presentation layer.
- Before writing
09-review.html, reusedesign-language.mdwhen it already exists and the user has not asked to refresh the visual direction. - When
design-language.mddoes not exist, usevisual-taste-labif available to analyze the PRD content, audience, density, risk profile, and handoff job, then create an appropriate design language. Do not ask the user to confirm style; choose a suitable direction and proceed. - If
visual-taste-labis unavailable, still createdesign-language.mdfrom the same content-first analysis and continue. Do not block PRD completion on visual style. - Do not make the HTML artifact a marketing page, fake product landing page, decorative dashboard, or visual demo unless the user explicitly asks.
- If Playwright or a browser tool is available, verify
09-review.htmlon desktop and mobile screenshots and fix visible overflow, clipped text, or incoherent layout before finishing.
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 · 107 lines · 1,305 tokens per session scan A 8a8edb970f1b
prd-agent-kit AGENTS.md is an instructions file published in the GitHub repository baobao2333/prd-agent-kit (5 stars, last pushed 3mo ago), licensed MIT. It adds 1,305 tokens to every session, about $0.0065 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 instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.