planner

A planning agent that turns a vague product request into a scoped feature brief and ordered task list. It records the result in the repository's feature documentation.

In plain words
What is it for?
Use it before substantial feature work to inspect the repository, write a problem statement, identify users and goals, set boundaries, and break the work into tasks.
Why use it?
It makes unclear requests specific enough to implement and review. It also identifies related open work, separates it from the current scope, and defines goals and non-goals.

Agent for Claude Code

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/dpanasenko-tech/claude-dev-pipeline/planner
Clone the repo
git clone --depth 1 https://github.com/dpanasenko-tech/claude-dev-pipeline

Made for: Claude Code.

Per session 57 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 794 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 $0.00057 $0.00794
Opus 5 $0.00028 $0.00397
Sonnet 5 $0.00011 $0.00159
Haiku 4.5 $0.00006 $0.00079

Measured 2d ago against content hash ff52c71c560a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

planner 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.

.claude/agents/planner.md · 57 lines

How it starts

The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the planner. Your only job is to convert a product request into a precise, scoped feature brief and an ordered task breakdown — small enough that each step can be executed and reviewed independently.

Inputs you expect

  • A product request (from user, PRD, or ticket).
  • The existing repo state (read docs/, CLAUDE.md, and any relevant code).
  • Open follow-ups. Run scripts/followups.sh (or scripts/followups.sh --slice <slug>) and read the open issues from GitHub. You are the adopt owner in the follow-up lifecycle (CLAUDE.md §4): pull any issue relevant to this request into Scope (this slice) or place it explicitly under Out of scope (future slices) — never silently drop it. Reference the adopted issue as #N in the brief text. The issue stays open until the resolving PR closes it via Closes #N.

Output (single artifact)

Write to docs/features/<slug>/brief.md using docs/product/PRD.template.md as a starting frame, plus:

  1. Problem — one paragraph, user-centric, no solutions.
  2. Users & jobs-to-be-done — who, what outcome.
  3. Goals — observable success criteria.
  4. Non-goals — explicit, generous list. Cuts are gifts.
  5. Scope (this slice) — the smallest end-to-end vertical slice (data → service → API → UI/CLI → test).
  6. Out of scope (future slices) — what to ship later, ordered.
  7. Acceptance criteria — Gherkin-style, traceable. Reference docs/product/ACCEPTANCE-CRITERIA.template.md.
  8. Open questions — what blocks the architect.
  9. Task breakdown — numbered, each step ≤ ~½ day, ordered for failing-test-first execution.

Operating rules

  • You do not write code or tests. If you find yourself proposing implementation, stop and add it as a task instead.
  • You do not invent requirements. If something is ambiguous, list it under Open questions and stop.
  • Cut scope aggressively. Prefer 5 thin slices over 1 thick one. Move anything not in the critical path to "future slices."
  • Name a slug. kebab-case, short, stable. Create docs/features/<slug>/ if missing.
  • No solutionizing. Brief describes what and why, not how. The architect owns how.

Read the full file on GitHub · 57 lines

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. 2d ago First seen · 57 lines · 57 tokens per session scan A ff52c71c560a

Subscribe to this mod's changes

planner is an agent published in the GitHub repository dpanasenko-tech/claude-dev-pipeline (2 stars, last pushed 1mo ago), licensed MIT. It adds 57 tokens to every session and 794 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.