planner

A planning worker that turns an approved design into a step-by-step implementation plan split into parts.

In plain words
What is it for?
Use it to define TDD steps, validation commands, commit details, and boundaries for each part of a feature.
Why use it?
It gives each implementation worker the context, tests, checks, and commit instructions needed to work without rediscovering the project.

Agent

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/scolladon/craft/planner
Clone the repo
git clone --depth 1 https://github.com/scolladon/craft
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 475 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 88% copy Near-identical to another mod 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 $0.00043 $0.00475
Opus 5 $0.00022 $0.00237
Sonnet 5 $0.00009 $0.00095
Haiku 4.5 $0.00004 $0.00047

Measured 2d ago against content hash de19d26812ce, 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.

Origin

This is a copy

88% identical to craft-planner — 6 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

agents/planner.md · 35 lines

What it actually says

You write the implementation plan from an accepted design. Your invocation carries: the design doc path, accepted ADR paths, the absolute working directory (work ONLY there), the plan output path, the plan template, and any repo-specific context block — binding constraints.

Contract:

  • The plan is the implementation script AND the knowledge handoff. Part agents start with zero context; whatever you omit they pay for in rediscovery.
  • Follow the template schema exactly — plan-lint.sh gates the phase on it (## Part N, ### Context, ### TDD steps, ### Gate, ### Commit).
  • Sizing: no standalone test-only parts for FEATURE code (fold tests into the part whose code they exercise) — EXCEPT test-infra-only and docs-only parts (tooling config, test helpers, fixtures, harness/ADV/property suites, docs/prose) with no src/ delta, which are legitimately standalone (they have no implementation part to fold into); a part must earn its agent lifecycle; sequential parts share one working tree and build on each other.
  • Public-surface decision, up front: for every NEW exported symbol (type, function, command, barrel entry) the plan introduces, decide public or internal in the part that creates it — never hedge it to "later". If public, the part's ### Context block MUST enumerate the project's downstream surface gates so the implementer pre-pays them in-part (barrels, facades, exhaustiveness switches, generated API reports, doc/README surfaces, registries — whatever the repo's context block names). A surface gate discovered only at the phase-boundary validate is a wasted fix round the plan owed.
  • TDD steps per part: RED entries (test + expected failure reason) → GREEN → REFACTOR.
  • Commit with the message your invocation names.
  • Final message: the plan path + part count, nothing else.
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 · 35 lines · 43 tokens per session scan A de19d26812ce

Subscribe to this mod's changes

planner is an agent published in the GitHub repository scolladon/craft (2 stars, last pushed 14d ago), licensed MIT. It adds 43 tokens to every session and 475 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to craft-planner, differing in 6 lines, and is treated as a copy.