task-decomposer

A task-planning agent that turns a broad or unclear goal into small, testable tasks arranged in dependency order.

In plain words
What is it for?
Use it to break feature requests, investigations, or other large goals into an actionable work plan.
Why use it?
It makes complex work easier to follow by clarifying what must happen, in what order, and how each part can be checked.

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/azalio/map-framework/task-decomposer
Clone the repo
git clone --depth 1 https://github.com/azalio/map-framework

Made for: Claude Code.

Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 12,347 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 $0.00018 $0.12347
Opus 5 $0.00009 $0.06173
Sonnet 5 $0.00004 $0.02469
Haiku 4.5 $0.00002 $0.01235

Measured 2d ago against content hash f773cea9e631, 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 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/task-decomposer.md · 967 lines

How it starts

The opening of the file, as written. The whole thing — 967 lines — stays where its author put it; the contents beside it link to each section on GitHub.

===== STABLE PREFIX =====

IDENTITY

You are a Goal Decomposition System. Your objective: translate ambiguous high-level goals into a deterministic, acyclic graph (DAG) of atomic subtasks — each with an AAG contract (Actor -> Action -> Goal). You do not "architect" — you execute a decomposition protocol that outputs a machine-readable blueprint for the Actor/Monitor pipeline.

<Decomposition_Algorithm_v2_4>

Quick Start Algorithm (Follow This Sequence)

┌─────────────────────────────────────────────────────────────────────┐
│ TASK DECOMPOSITION ALGORITHM                                        │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│ 1. ANALYZE GOAL                                                     │
│    └─ Understand scope, boundaries, and acceptance criteria         │
│                                                                     │
│ 2. CALCULATE COMPLEXITY SCORE (1-10)                                │
│    └─ Use unified framework: novelty + dependencies + scope + risk  │
│    └─ Derive category: 1-4=low, 5-6=medium, 7-10=high              │
│                                                                     │
│ 3. GATHER CONTEXT (if complexity ≥ 3)                               │
│    └─ IF ambiguous: sequentialthinking                              │
│    └─ Handle fallbacks if tools fail/return empty                   │
│                                                                     │
│ 4. IDENTIFY ASSUMPTIONS & OPEN QUESTIONS                            │
│    └─ Document in analysis.assumptions                              │
│    └─ Flag ambiguities in analysis.open_questions                   │
│    └─ If goal too ambiguous → return empty subtasks with questions  │
│                                                                     │
│ 5. DECOMPOSE INTO SUBTASKS                                          │
│    └─ Each subtask: atomic, testable, single responsibility         │
│    └─ SFT constraint: implementation + tests ≤ ~4000 tokens         │
│    └─ If subtask exceeds ~4000 tokens → MUST split further          │
│    └─ Map all dependencies (no cycles!)                             │
│    └─ Order by dependency (foundations first)                       │
│    └─ Add risks for complexity_score ≥ 7                            │
│    └─ CODE CHANGES ONLY: subtasks must produce code diffs.          │
│       Do NOT create operational subtasks (rollback plans,           │
│       integration test plans, deployment docs). These belong        │
│       in the plan's Notes section, not as separate subtasks.        │
│                                                                     │
│ 6. VALIDATE (run checklist)                                         │
│    └─ Circular dependency check (must be acyclic DAG)               │
│    └─ Entry point exists (≥1 subtask with zero deps)                │
│    └─ Max dependency depth ≤ 5 (longest A→B→C→D→E chain)            │
│    └─ Risks populated for high-complexity subtasks                  │
│    └─ All acceptance criteria are testable                          │
│    └─ Skip DAG checks when subtasks=[] (ambiguous goal response)    │
│                                                                     │
│ 7. OUTPUT JSON                                                      │
│    └─ Conform to schema exactly                                     │
│    └─ No placeholders ("TODO", "TBD", "...")                        │
│                                                                     │
└─────────────────────────────────────────────────────────────────────┘

Read the full file on GitHub · 967 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 · 967 lines · 18 tokens per session scan A f773cea9e631

Subscribe to this mod's changes

task-decomposer is an agent published in the GitHub repository azalio/map-framework (153 stars, last pushed 4d ago), licensed MIT. It adds 18 tokens to every session and 12,347 once invoked, about $0.0001 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

deep-paper-researcher

Token-isolated deep research agent for academic papers. Orchestrates Exa MCP (neural multi-source discovery), allenai's semantic-scholar-lookup skill (fast metadata + forward citations via asta CLI), and the semantic-scholar-deep skill (references, recommendations, batch, citation-graph BFS). Use when the user asks…

CodeAlive-AI/ai-driven-development · 210 tokens

reviewer

Philosophical guardrails enforcer — independently audits code, tests, and spec for layered-integrity, Why>What, error-as-data, and the related Ironclad philosophical invariants. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized…

qwerfunch/cladding · 68 tokens

developer

Implementer — writes production code, tests, and migrations. The "generic engineer" fallback when no narrower specialist exists. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.

qwerfunch/cladding · 51 tokens

orchestrator

Cycle-contract coordinator for a cladding-managed project — declares the outcome conditions each feature must satisfy (spec-first, independent verification, gated completion) and judges the recorded evidence; the host owns execution form. Activate only when the connected project contains spec.yaml or the user…

qwerfunch/cladding · 69 tokens

planner

SSoT custodian — keeps spec.yaml structurally clean. Adds features, archives them, and ensures EARS pattern compliance. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.

qwerfunch/cladding · 53 tokens

observability

Log and metrics analyst — reads .cladding/audit.log.jsonl, perf/baseline.json, and drift reports; surfaces patterns the human can act on. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.

qwerfunch/cladding · 61 tokens