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/sananthanarayan/skilldrop/prd-draftnpx skills add sananthanarayan/skilldrop --skill prd-draftgit clone --depth 1 https://github.com/sananthanarayan/skilldropWhat 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 | $0.00082 | $0.01438 |
| Opus 5 | $0.00041 | $0.00719 |
| Sonnet 5 | $0.00016 | $0.00288 |
| Haiku 4.5 | $0.00008 | $0.00144 |
Grade A, and why
prd-draft 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
prd-draft
Turns "the business wants X" into the document that aligns everyone on what problem, for whom, how we'll know — before anyone argues about how. The missing link in the pipeline: brief-intake (raw mess → brief) → prd-draft → user-story-splitter (stories), nfr-spec (quality targets), success-metrics (measurement), design-doc (the how).
How to respond
-
Ingest the idea — a
brief-intakeoutput, a pitch paragraph, a meeting note. Ask at most 2 questions, spent on the two highest-leverage unknowns: evidence ("what tells us users actually have this problem?") and the binding constraint ("fixed deadline, fixed scope, or fixed team?"). Everything else: pick a default, tag it[assumption]. -
Write the problem statement with zero solution nouns. It names users, their situation, the pain, and the evidence — and survives the test: could this paragraph justify a completely different solution than the one everyone has in mind? ✅ "Support agents spend ~20 min/ticket reconstructing customer order history across three tools
[reported by support lead]" — ❌ "We need an order-history dashboard" (that's a solution wearing a problem costume). -
Name the users specifically enough to find one. Primary persona + their job-to-be-done; secondary personas listed but explicitly deprioritized. "All users" is not a persona — if the feature really serves everyone, name who feels the pain most.
-
State goals with a success line each. Every goal carries one measurable "we'll know it worked when …" sentence — target, timeframe, baseline if known. One line here; the full measurement design (instrumentation, guardrails) is
success-metrics' job — point the user there. -
Write requirements that are testable and solution-free, each with a MoSCoW priority — and the Won't-have list is mandatory: ✅ "M — Agent sees a customer's orders from all channels in one view, ≤3s after lookup" — ❌ "M — Use Redis to cache order data" (implementation, belongs in
design-doc) — ❌ "S — The experience is seamless" (untestable, belongs nowhere). Requirements describe observable behavior or capability; each one is checkable by a tester or demo.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 63 lines · 82 tokens per session scan A 8ff1cd74eeb8
prd-draft is a skill published in the GitHub repository sananthanarayan/skilldrop (2 stars, last pushed 18d ago), licensed MIT. It adds 82 tokens to every session and 1,438 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 skills, from other repositories
openlore-brainstorm
Transform a feature idea into an annotated story using a Domain Sketch or Constrained Option Tree. Use when asked to brainstorm, explore, or shape a feature before implementation.
openlore-analyze-codebase
Run a full static OpenLore analysis and summarize architecture, call graph, refactoring issues, and duplicate code. Use when asked to analyze, map, or assess a codebase without LLM inference.
openlore-review-changes
Review code changes using OpenLore risk, call, coverage, and cluster evidence without editing code. Use when asked for a change review, pre-PR safety check, or merge recommendation.
architect-implement
Generate a full Architecture Description (AD.md) from accepted ADRs using multi-agent DAG orchestration. Use when accepted ADRs exist and you need to produce or update unified architecture documentation.
team-repair
Re-index OKF v0.2 index.md/log.md files, derive CDR.md, rebuild .skills.json and AGENTS.md in team-ai-directives, migrate v0.1→v0.2 frontmatter, scan for rule conflicts, and verify directive freshness. Use when indexes are inconsistent, orphans are detected, after bulk changes, or for periodic team AI directives…
mission-brief
Mission-driven SDD orchestrator: take a feature description, structure it into a Mission Brief (goal, constraints, success criteria), generate an ordered step list with prompts that trigger installed SDD skills via model invocation or command-file discovery, and walk those steps to converged implementation. Use when…