task-decomposer

task-decomposer is an agent for Claude Code from rjmurillo/ai-agents. It costs 84 tokens per session (2,458 once invoked), scanned A, original, MIT.

A task-planning agent that turns product requirements documents and large project work into small, ordered tasks with acceptance criteria. Acceptance criteria are checks that show when a task is finished.

In plain words
What is it for?
Use it to break down PRDs or epics, identify dependencies, group milestones, estimate complexity, and define concrete done conditions.
Why use it?
It makes broad or unclear work easier to estimate, assign, sequence, and verify.

Agent for Claude Code

Part of the project-toolkit plugin — 22 commands, 33 agents shipped together

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/rjmurillo/ai-agents/task-decomposer
Clone the repo
git clone --depth 1 https://github.com/rjmurillo/ai-agents

Made for: Claude Code.

Or install project-toolkit, the plugin that ships this one along with the rest of its 22 commands, 33 agents.

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 task-decomposer

README.md
[![agentmods](https://agentmods.dev/badge/agents/rjmurillo/ai-agents/task-decomposer.svg)](https://agentmods.dev/agents/rjmurillo/ai-agents/task-decomposer)
Your own site
<a href="https://agentmods.dev/agents/rjmurillo/ai-agents/task-decomposer"><img src="https://agentmods.dev/badge/agents/rjmurillo/ai-agents/task-decomposer.svg" alt="Measured on agentmods" height="20"></a>
Per session 84 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,458 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 $0.00084 $0.02458
Opus 5 $0.00042 $0.01229
Sonnet 5 $0.00017 $0.00492
Haiku 4.5 $0.00008 $0.00246

Measured 4d ago against content hash 77c119399a75, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-decomposer 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 4d 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/task-decomposer.md · 326 lines

How it starts

The opening of the file, as written. The whole thing — 326 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Task Decomposer

Core Identity

Task Decomposition Specialist breaking PRDs and epics into atomic, estimable work items.

Style Guide Compliance

Key requirements:

  • No sycophancy, AI filler phrases, or hedging language
  • Active voice, direct address (you/your)
  • Replace adjectives with data (quantify impact)
  • No em dashes, no emojis
  • Text status indicators: [PASS], [FAIL], [WARNING], [COMPLETE], [BLOCKED]
  • Short sentences (15-20 words), Grade 9 reading level

Agent-Specific Requirements:

  • Quantified task estimates: Use complexity sizes (XS/S/M/L/XL) with clear guidelines
  • Clear acceptance criteria format: Verifiable checkboxes, not vague descriptions
  • Evidence-based estimates: Include reconciliation when derived estimates diverge >10%
  • Text status indicators: Use [PASS], [FAIL], [PENDING] instead of emojis
  • Active voice: "Implement the feature" not "The feature should be implemented"

Activation Profile

Keywords: Decomposition, Atomic-tasks, Breakdown, Acceptance-criteria, Complexity, Estimates, Dependencies, Sequencing, Milestones, Work-items, TASK-ID, Assignable, Trackable, Boundaries, Discrete, Done-criteria, Reconciliation, Phases, Verification, Scope

Summon: I need a task decomposition specialist who breaks PRDs and epics into atomic, estimable work items with clear acceptance criteria and done definitions. You sequence by dependencies, group into milestones, and size by complexity, not time. Each task should be discrete enough that someone can pick it up and know exactly what to do. Reconcile estimates and flag scope concerns before they become problems.

Claude Code Tools

You have direct access to:

  • Read: PRDs and existing code
  • Grep/Glob: Find relevant files
  • TodoWrite: Track generation progress
  • Bash: gh issue create for GitHub issues
  • Memory Router (ADR-037): Unified search across Serena + Forgetful
    • uv run python .claude/skills/memory/scripts/search_memory.py --query "topic"
    • Serena-first with optional Forgetful augmentation; graceful fallback
  • Serena write tools: Memory persistence in .serena/memories/
    • mcp__serena__write_memory: Create new memory
    • mcp__serena__edit_memory: Update existing memory

Read the full file on GitHub · 326 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. 4d ago First seen · 326 lines · 84 tokens per session scan A 77c119399a75

Subscribe to this mod's changes

task-decomposer is an agent published in the GitHub repository rjmurillo/ai-agents (45 stars, last pushed today), licensed MIT. It adds 84 tokens to every session and 2,458 once invoked, about $0.0004 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.

Related

Other agents, from other repositories

code-reviewer

Use this agent for code review before commits or PRs. Checks for bugs, security issues, performance problems, and code quality. Provides actionable feedback with specific line references.

arpitnath/claude-capsule-kit · 40 tokens

database-navigator

PROACTIVELY use this agent when exploring database schemas, understanding data models, or investigating database-related issues. Expert in SQL, migrations, and data relationships. Read-only agent for production safety.

arpitnath/claude-capsule-kit · 43 tokens

security-analyst

Security vulnerability analyst for authentication, authorization, secrets, input validation. Use proactively when code changes involve auth/security patterns or when explicitly requested via /review security. Activates for auth/, security/, middleware/ paths and keywords: auth, jwt, oauth, session, password, secret…

changoo89/claude-pilot · 0 tokens

devops-engineer

CI/CD, infrastructure, and deployment specialist for Solana projects. Handles GitHub Actions, Docker, monitoring, RPC management, and Cloudflare Workers edge deployment. Use when: Setting up CI/CD pipelines, containerizing Solana validators or programs, configuring monitoring and alerting, managing RPC infrastructure…

solanabr/solana-vault-standard · 76 tokens

epidemiologist

Use to challenge, audit, or design epi methods. Triggers on: 'review my study design', 'is this valid', 'what are the biases', 'case definition', 'surveillance system', 'outbreak investigation', 'is the denominator right', 'PPV in a low-prevalence setting', 'SaTScan'…

SVerITG/Metis_PH · 0 tokens

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens