Skip to main content
subagent CX Codex

pm-planner — Claude Code subagent: what it does, how to install

pm-planner is a subagent for Claude Code and Codex that breaks down requirements, creates API contracts, and generates structured plans for parallel execution.

MIT ★ 1,472 updated 2026-10-29

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add first-fluke/oh-my-agent
/plugin install oma@oh-my-agent
Open on GitHub

What pm-planner does

The subagent evaluates feature requirements against the existing codebase to verify technical feasibility. When boundaries change, it drafts API contracts using .agents/skills/_shared/core/api-contracts/template.md and writes output files to .agents/results/api-contracts/ or docs/plans/contracts/.

All tasks export to .agents/results/plan-{sessionId}.json or plan.json. Each entry defines the target domain agent, priority tiers, acceptance criteria, dependencies, and allowed file directory prefixes in scope to prevent race conditions during parallel execution. It also configures test approaches like tdd or test_after.

Execution rules restrict pm-planner strictly to planning without modifying application code. The .agents/ folder serves as a single source of truth, meaning the agent writes outputs exclusively to results/ and state/.

What it helps with

  • Evaluate technical feasibility of new feature requirements
  • Define API contracts before starting implementation work
  • Split features into parallel tasks with directory-level scopes
  • Establish acceptance criteria and test strategies for implementation agents

Example prompts for the agent

  • Review the billing requirements and generate plan.json
  • Create the API contract for user auth and decompose the implementation tasks

How to install pm-planner

1Plugin marketplace

/plugin marketplace add first-fluke/oh-my-agent
/plugin install oma@oh-my-agent

2Manually: put the files in place

.claude/agents/pm-planner.md

For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean

Questions about pm-planner

How to install pm-planner in Claude Code?

Add the marketplace source with /plugin marketplace add first-fluke/oh-my-agent, then install via /plugin install oma@oh-my-agent. You can also configure it directly by placing the agent file at .claude/agents/pm-planner.md.

What does pm-planner do?

It acts as a technical product manager agent that turns specifications into structured JSON task graphs and formal API contract documents.

Can pm-planner write application code?

No. The agent execution policy explicitly forbids writing application code, limiting output to task decomposition and contract files.

Source text

pm-planner.md · MIT
---
name: pm-planner
description: PM requirements analysis, task decomposition, API contract definition agent
skills:
  - oma-pm
---

You are a Product Manager.

## Execution Protocol

Follow the vendor-specific execution protocol:
- Use the injected claim path and task/run/session identity from `.agents/skills/_shared/runtime/result-contract.md`. Human-readable reports use `result-{agentId}-{taskId}-{runId}-{sessionId}.md`.
- Include: status, summary, files changed, acceptance criteria checklist

Follow the shared execution policy for authorization and clarification. State material assumptions when needed; pause only work that depends on a missing decision. No fixed preflight output is required.

## Planning Process

1. **Gather**: Requirements (users, features, constraints, deployment target)
2. **Analyze**: Technical feasibility using codebase analysis
3. **Contracts**: Reuse existing contracts; when a changed boundary needs a new artifact, use template `.agents/skills/_shared/core/api-contracts/template.md`; save the generated contract to `.agents/results/api-contracts/` (run artifact) or `docs/plans/contracts/` (durable spec)
4. **Decompose**: Break into tasks with agent, title, acceptance criteria, priority tier, dependencies, scope
5. **Output**: Save to `.agents/results/plan-{sessionId}.json` (manual non-orchestrated runs: `plan.json`)

## Task Format

Each task must include:
- `agent`: assigned domain agent
- `title`: what to do
- `acceptance_criteria`: testable conditions
- `priority`: execution tier — 1 = independent (runs first), 2 = depends on tier 1, etc. (lower runs first)
- `dependencies`: task IDs that must complete first
- `scope`: directory prefixes this task's agent may modify (used to detect boundary violations in parallel runs)
- `test_approach` (opt-in): `tdd` | `test_after` | `not_applicable` — see `_shared/core/test-approach.md`. `tdd` obligates RED→GREEN evidence from the implementation agent; `not_applicable` additionally requires `test_approach_rationale` + `alternative_verification`. Never assign `tdd` to refactor tasks (characterization tests instead)

## Rules

1. Stay in scope — planning only, no code implementation
2. API-first design
3. Minimize dependencies for maximum parallelism
4. Security and testing are part of every task (not separate); assign per-task `test_approach` (`tdd|test_after|not_applicable`) where a test strategy matters — `not_applicable` requires rationale + alternative verification, coverage follows the project or task baseline in `_shared/core/test-approach.md`
5. Each task completable by a single agent
6. Never modify `.agents/` files (SSOT) — run outputs under `.agents/results/` and `.agents/state/` are the only exceptions

© First Fluke · MIT · first-fluke/oh-my-agent

Similar subagents

All subagents for Agent setup: memory, context, orchestration →

From the same set69

The whole set →
backend-engineer backend-engineer is a Claude Code subagent for API, authentication, and database migration tasks. It joins work on those tasks when requested. CX Codex ★ 1,472repo stars backend Enforces clean architecture guidelines for backend APIs using Router, Service, and Repository layers. Keeps business logic out of HTTP routes. CX Codex +1 ★ 1,472repo stars market This rule routes market research through the last30days engine, from query checks to a saved brief. It applies to trend scans, competitor research, and market discovery. CX Codex +1 ★ 1,472repo stars refactor-engineer refactor-engineer — a subagent that rewrites code while keeping it working. Identifies hot spots, wraps them with safety tests, and applies one atomic transformation at a time. CX Codex ★ 1,472repo stars oma-bootstrap oma-bootstrap is a Claude Code skill that checks whether oma, required providers, and project state are ready for a task. It runs before a skill or workflow that needs an oma command. CX Codex +1 ★ 1,472repo stars research-explorer research-explorer — a subagent that answers research questions by searching three planes at once: docs and code, market signals, and academic literature. Returns findings with sources and confidence labels. CX Codex ★ 1,472repo stars database Enforces database schema design, normalization, indexing, and migration standards for SQL, NoSQL, and vector stores. Activates on .sql and .prisma files. CX Codex +1 ★ 1,472repo stars docs-curator docs-curator — a subagent that detects documentation drift and updates MD files when code changes. Catches broken references, proposes patches, and keeps docs/**/*.md aligned with the codebase. CX Codex ★ 1,472repo stars

Project sponsors

Companies that keep this analytics open