task:decompose

A command for breaking a large task into smaller subtasks, each intended to take no more than roughly five to eight hours.

In plain words
What is it for?
Use it to turn broad development work into a sequence of manageable tasks while preserving task-status rules.
Why use it?
Smaller tasks are easier to schedule, assign, track, and retry when part of the work is blocked.

Command 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 commands/xsaven/vector-memory-mcp/decompose
Clone the repo
git clone --depth 1 https://github.com/Xsaven/vector-memory-mcp

Made for: Claude Code.

Per session 20 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 8,132 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 $0.00020 $0.08132
Opus 5 $0.00010 $0.04066
Sonnet 5 $0.00004 $0.01626
Haiku 4.5 $0.00002 $0.00813

Measured yesterday against content hash aa6f82d357c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task:decompose 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 yesterday.

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/commands/task/decompose.md · 370 lines

How it starts

The opening of the file, as written. The whole thing — 370 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Iron Rules

Status-semantics (CRITICAL)

Task status has STRICT semantics: "pending" = waiting to be worked on (includes failed/blocked tasks returned to queue). "in_progress" = currently being worked on. "completed" = implementation done, ready for validation. "tested" = tests written/passed. "validated" = passed all quality gates. "stopped" = PERMANENTLY CANCELLED — task is NOT needed, will NEVER be executed. ONLY set "stopped" when: user explicitly requests cancellation, OR task is provably unnecessary (duplicate, superseded, irrelevant). NEVER set "stopped" for: failures, blocks, validation issues, tool errors, missing dependencies. For these → set "pending" with detailed blocker in comment.

  • why: Agents misuse "stopped" as "failed/blocked" which breaks workflow permanently. A stopped task is removed from pipeline — it will never be picked up again. A pending task with a blocker comment will be retried, either automatically or manually.
  • on_violation: If about to set "stopped": verify it is a TRUE cancellation. If task failed or is blocked → set "pending" + comment explaining what happened. "stopped" is irreversible workflow termination.

Task-get-first (CRITICAL)

FIRST TOOL CALL = mcp__vector-task__task_get. No text before. Load task, THEN analyze how to decompose.

No-hallucination (CRITICAL)

NEVER output results without ACTUALLY calling tools. You CANNOT know task status or content without REAL tool calls. Fake results = CRITICAL VIOLATION.

Read the full file on GitHub · 370 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. yesterday First seen · 370 lines · 20 tokens per session scan A aa6f82d357c5

Subscribe to this mod's changes

task:decompose is a command published in the GitHub repository Xsaven/vector-memory-mcp (0 stars, last pushed 6mo ago), licensed MIT. It adds 20 tokens to every session and 8,132 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 2026-09-01.