A specification-drafting helper that turns a product requirements document (PRD) into a reviewable technical plan for one code repository. It writes the plan locally and prepares information for opening a draft pull request.

Use it to identify relevant capabilities, write structured requirements and acceptance criteria, trace them back to the PRD, and list questions for developer review before implementation.

It removes the gap between product language and the engineering details needed by developers. It also limits the plan to the current repository and highlights unclear requirements instead of guessing.

spec-draft — Skill for Claude Code by drivestream-lab

Install

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.

Any agent

npx skills add drivestream-lab/prayog-skills --skill spec-draft

Clone the repo

git clone --depth 1 https://github.com/drivestream-lab/prayog-skills

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/drivestream-lab/prayog-skills/spec-draft/github.svg)](https://agentmods.dev/skills/drivestream-lab/prayog-skills/spec-draft)

Your own site

<a href="https://agentmods.dev/skills/drivestream-lab/prayog-skills/spec-draft"><img src="https://agentmods.dev/badge/skills/drivestream-lab/prayog-skills/spec-draft/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/drivestream-lab/prayog-skills/spec-draft"><img src="https://agentmods.dev/badge/skills/drivestream-lab/prayog-skills/spec-draft.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 111 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,539 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00111	$0.03539
Opus 5	$0.00056	$0.01769
Sonnet 5	$0.00022	$0.00708
Haiku 4.5	$0.00011	$0.00354

Measured yesterday against content hash a0dba5907a84, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

spec-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 yesterday.

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.

skills/development/spec-draft/SKILL.md · 253 lines

How it starts

The opening of the file, as written. The whole thing — 253 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Spec draft

Translate the PRD into a spec slice for this repo. The spec slice is the engineering team's interpretation of what the PRD means for their codebase.

Do not implement. Produce docs/specification/product/INIT-{id}.md only.

Why this skill exists

PM writes PRD in feature language. Engineers own the spec. This skill bridges the gap — it reads the PRD and drafts structured spec REQ-* rows that the dev team can review, edit, and then run /initiative-feasibility on.

NON-NEGOTIABLE

The spec draft is a starting point — dev must review and edit before committing. Do not present it as authoritative without dev review.
Language in the spec must be engineering terms — REQ-*, acceptance criteria, module scope — not copied PRD user-story language. Ids follow prayog-skills/references/id-conventions.md (REQ-* canonical; legacy FR-* ≡ same number).
Scope must be bounded to this repo only. Do not write REQs for other repos.
Every REQ-* must trace to a named PRD CAP-* / REQ-* or section/bullet.
Flag anything in the PRD that is ambiguous for this repo — do not guess. Put ambiguities in a "Spec questions" section at the bottom.
Handoff gate first (Gate 1 / G1 + H1–H3). The canonical impact-map artifact, source PRD digest (H1), current meta PR head SHA, and tech-lead APPROVED review must match. A label or old LGTM alone is not approval. Open IM-{nn} with Blocking: yes fails closed (D1b) — do not draft the spec until resolved or deferred with default.
This repo must be affected in the latest map and not deferred or blocked. Record the repo's scope_digest (H2) and map revision (H3) in the spec header (H4 citations). If any gate fails: stop.
No forge mutations in this skill. Do not create a branch, commit, push, PR, comment, review request, apply labels, create issues, or merge here. Persist the spec locally and fill handoff.forge readiness for open_draft_pr; recommend /open-draft-pr or /commit-workspace (or wait for the orchestrator ForgeClient on spec-pr-action). See prayog-skills/references/forge-side-effects.md#content-producers.
Treat spec-* labels as projections. If artifact, review, and label disagree, the gate is closed. Exactly one PE gate label may be active: spec-pending, spec-lgtm, or spec-blocked. spec-revised and spec-stale are additional invalidation labels and always close the gate.
Ownership boundary. The spec owns observable product behavior, acceptance, field meaning, invariants, errors, and compatibility. It may cite existing architectural constraints from adr_dir but must not choose implementation architecture. Architecture questions are recorded for feasibility / technical review — not decided here.
Outcome vocabulary. Emit only pass, needs-input, blocked, stale, or failed (this stage has no findings edge). Map evidence to the outcome rubric below; do not invent lane-specific outcomes.
If a codegraph provider is available (MCP tool matching prayog-skills/references/codegraph-tool-contract.md, or a local CLI), prefer it for architecture/impact questions. Always fall back to direct source_roots reads when unavailable — never block or change outcome selection on its absence.
Carry Capabilities and Journeys. In-scope PRD CAP-* and J-{nn} appear in the spec tables (or explicit out-of-repo N/A). Do not drop journeys silently. Live evidence REQs cite CAP-* and/or J-* when the PRD defines those namespaces (quality-confidence-ladder.md).

Read the full file on GitHub · 253 lines

Files

What ships with it

7 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.

Changes

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.

yesterday Changed · +5 lines a0dba5907a84
9d ago First seen · 248 lines · 111 tokens per session scan A 0adfa3c14622

Subscribe to this mod's changes

spec-draft is a skill published in the GitHub repository drivestream-lab/prayog-skills (2 stars, last pushed yesterday), licensed MIT. It adds 111 tokens to every session and 3,539 once invoked, about $0.0006 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.

Planning Writing and docs

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens