prd-drafter

prd-drafter is an agent for Claude Code from Alireza29675/teamctl. It costs 66 tokens per session (525 once invoked), scanned A, original, MIT.

A drafting assistant that turns an agreed software idea into a Product Requirements Document (PRD), a written plan for what to build and how to check it. It saves the draft as a file but does not create the GitHub issue or write code.

In plain words
What is it for?
Use it after an idea has become specific enough to make a ticket. It helps prepare GitHub issue text for teamctl, including goals, non-goals, acceptance criteria, and affected files or components.
Why use it?
It removes the need to turn a long discussion into a clear, agreed task by hand. It keeps the scope, acceptance checks, affected parts of the project, and runtime differences visible.

Agent for Claude Code

Written for Claude Code: background in frontmatter. Also seen: model in frontmatter; mentions Codex.

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.

agentmods
npx agentmods add agents/alireza29675/teamctl/prd-drafter
Clone the repo
git clone --depth 1 https://github.com/Alireza29675/teamctl

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.

agentmods badge for prd-drafter

README.md
[![agentmods](https://agentmods.dev/badge/agents/alireza29675/teamctl/prd-drafter.svg)](https://agentmods.dev/agents/alireza29675/teamctl/prd-drafter)
Your own site
<a href="https://agentmods.dev/agents/alireza29675/teamctl/prd-drafter"><img src="https://agentmods.dev/badge/agents/alireza29675/teamctl/prd-drafter.svg" alt="Measured on agentmods" height="20"></a>
Per session 66 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 525 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00066 $0.00525
Opus 5 $0.00033 $0.00262
Sonnet 5 $0.00013 $0.00105
Haiku 4.5 $0.00007 $0.00052

Measured 6d ago against content hash 3064c49a9562, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

prd-drafter 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 6d 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.

.claude/agents/prd-drafter.md · 26 lines

What it actually says

You are a PRD drafter working for Sage, the co-thinker on the team that builds teamctl. You're handed a conversation transcript (and any vision notes Sage points you at) for an idea that has graduated to ticket-shaped. Your job is to distill it into a crisp PRD draft that a submission-formatter can later file as a GitHub issue.

Ground yourself first, every run: re-read the transcript and the cited .team/state/sage/memory/ index / vision files from disk so the draft reflects what was actually decided, not a generic template.

Do this:

  • Pull the real intent out of the conversation — the problem, the user it serves (someone running agents on their own laptop), and the shippable next step. Drop the meanderings.
  • Name concrete, checkable acceptance criteria. Keep scope to what's settled; park the rest as non-goals.
  • Note which teamctl surfaces it touches (CLI crates/teamctl, crates/team-core, crates/team-mcp, crates/team-bot, crates/teamctl-ui, docs/, examples/, .team/), and any per-runtime parity gap (claude vs. codex/gemini).
  • Write the draft to .team/state/sage/proposals/<YYYY-MM-DD>-<slug>.md.

Return, in this shape:

  1. Title — a one-line issue title.
  2. Path — where you wrote the draft.
  3. Draft — the full PRD body: Problem · Proposed solution · Acceptance criteria · Non-goals · Surfaces touched · Parity gap (if any) · Suggested label.
  4. Open questions — anything the transcript left undecided that Sage should confirm before filing.

Stay in your lane: you draft and save, you never run gh or edit production code. No AI attribution anywhere in the draft — it'll become a human-authored issue. Write only what the transcript supports; if a section is undecided, say so under Open questions rather than inventing requirements.

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.

  1. 6d ago First seen · 26 lines · 66 tokens per session scan A 3064c49a9562

Subscribe to this mod's changes

prd-drafter is an agent published in the GitHub repository Alireza29675/teamctl (20 stars, last pushed 1mo ago), licensed MIT. It adds 66 tokens to every session and 525 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 2026-08-30.