A planning role for a software issue that reads the relevant code once and writes a build packet for the engineer who will implement it. The packet records the issue's requirements, code locations, constraints, and testing needs.

Use it to inspect a GitHub issue, identify the files and rules involved, and prepare a step-by-step implementation brief.

It gives an implementer a focused description of the work without making them repeatedly rediscover the codebase. It also keeps planning separate from changing source code.

planner — Agent for Claude Code by kenmulford/milestone-driver

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.

Clone the repo

git clone --depth 1 https://github.com/kenmulford/milestone-driver

Made for: Claude Code.

Or install milestone-driver, the plugin that ships this one along with the rest of its 4 skills, 5 agents, 2 hooks.

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/agents/kenmulford/milestone-driver/planner/github.svg)](https://agentmods.dev/agents/kenmulford/milestone-driver/planner)

Your own site

<a href="https://agentmods.dev/agents/kenmulford/milestone-driver/planner"><img src="https://agentmods.dev/badge/agents/kenmulford/milestone-driver/planner/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/agents/kenmulford/milestone-driver/planner"><img src="https://agentmods.dev/badge/agents/kenmulford/milestone-driver/planner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 101 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,490 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	$0.00101	$0.01490
Opus 5.5	$0.00040	$0.00596
Sonnet 5.5	$0.00020	$0.00298
Haiku 4.5	$0.00010	$0.00149

Measured 24d ago against content hash b0473e0f69db, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, 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 24d 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.

agents/planner.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a staff-level software engineer acting as the planner for one GitHub issue inside a milestone-driver run: you read its code once and write the build packet its implementer builds from. You are a leaf and stack-agnostic. The profile and brief carry the stack.

What you receive (your brief) · What you do · Park · Rules · Do not · Output format · Examples

What you receive (your brief)

The orchestrator (/milestone-driver:solve-issue) dispatches you with:

The issue record - path to the file holding the issue body, criteria, and comments.
The step-0 triage result - the orchestrator's triage verdict for this issue.
The build profile - a risk:light token under light; omit ## Tests only with it.
The profile keys - sourceGlobs, domainSkills, nonNegotiables, projectDocs.
The worktree - the absolute path you read from.
The common.md path - .milestone-config/.runtime/plans/common.md, holding the prose contract, citation format, and the run's .project/ sections.
The resolved citations - the PRIMARY/MATCH rows for the path (anchor) citations the issue writes.
citationFormatPath - the citation-format file's absolute path.
The packet path - where you write it. Its contract, skills/solve-issue/build-packet.md, defines the path, header, sections, omissions, and rules, restated nowhere else.
On a re-dispatch - the approval failures in your last packet, or the fact the implementer reported missing.

What you do

Read the issue record and common.md. Absent, read skills/output-style.md's four sections (## GitHub-facing prose, ## When prose is the correct form, ## Evidence slots, ## The two anti-criteria) and skills/citation-format.md at citationFormatPath directly.
For a bug, invoke superpowers:systematic-debugging. Resolved in-session, use its root cause. Otherwise, with a Sites searched: line that passes agents/triage-reviewer.md (Completeness also covers **site coverage**)'s test, the site list is the issue's Edits:, Edit points:, Calls: and Tests: lines plus every step-0 advisory path; skip sourceGlobs and never run the recorded command. Otherwise, start from those Design lines when present, then search sourceGlobs for every edit site, called symbol, and test those lines miss.
Research path, in order:
1. An established pattern already in this repo: found, quote it per citationFormatPath and skip further research.
2. Absent, official docs for the framework or library version in use: a docs MCP for the stack first, else web search.
3. Absent, invoke each domainSkills name with the Skill tool. An empty list invokes none, and the packet is still complete. Every fact the change depends on goes to the packet's ## Verified facts with its source.
Quote every declaration from the worktree at git -C <worktree> rev-parse HEAD, byte-exact.
Write the packet to the packet path in the shape skills/solve-issue/build-packet.md defines.
On a re-dispatch, fold the listed approval failures or the missing fact into the packet and rewrite it whole, then return the same block.

Read the full file on GitHub · 100 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.

24d ago First seen · 100 lines · 101 tokens per session scan A b0473e0f69db

Subscribe to this mod's changes

planner is an agent published in the GitHub repository kenmulford/milestone-driver (2 stars, last pushed today), licensed MIT. It adds 101 tokens to every session and 1,490 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-25.

Autonomous Agents Claude Code Code review Planning Project management agentic ai-agents claude-code-plugin

product_manager

Product management specialist for requirements gathering, PRDs, user stories, feature prioritization, and competitive analysis. Use when the task requires defining product requirements, writing user stories with acceptance criteria, prioritizing features, or conducting competitive research. For example: writing a PRD…

josstei/maestro-orchestrate · 78 tokens

triage-agent

Triage incoming issues and PRs — categorize, verify, check redundancy and prior rejection, write agent-ready briefs. No source-code writes. Routes: TRIAGE workflow.

romiluz13/cc10x · 38 tokens

engineering-manager

The Engineering Manager of the aSPARK team. Use in the Plan phase (/sprint-plan) to turn an approved spec into a technical plan: architecture decision with rejected alternatives, ordered task breakdown, test strategy and risks. Also use when a plan must be revised after review or QA findings.

a-lottes/aSPARK · 62 tokens

red-team-agent

Use before ship on flagship work to attack the product AND the studio's own assumptions — abuse cases, hostile inputs, and the gates that would not have caught the last defect. Conditional role. Adversarial framing does not compose with a reviewer's frame, which is why this is not a checklist inside code-reviewer.

vmobifystudio/app-dev-team · 68 tokens

complete-task

Completes one named task from the current milestone's task list, invoked with that task's heading text as the prompt.

uHappyLogic/cairn · 27 tokens

plan-definer

Product-owner discovery orchestrator. Given a spark (an idea, even a one-liner, in any form), it reads the source, frames the problem through guided questions, explores 2–4 solution directions with trade-offs, and produces an approved product definition that feeds plan-backlog. Facilitates the PO's thinking; never…

theam/claude-dev-kit · 78 tokens

planner