task-generator

An agent that converts Product Requirements Documents (PRDs) into structured implementation tasks managed by Claude Code.

In plain words
What is it for?
Use it to analyze a PRD, create atomic tasks, infer blocking relationships, show a proposed summary for approval, and merge with existing tasks.
Why use it?
It removes the need to manually interpret requirements, split them into work, and map which tasks must happen first.

Agent

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/sequenzia/claude-plugins/task-generator
Clone the repo
git clone --depth 1 https://github.com/sequenzia/claude-plugins
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,696 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 $0.00017 $0.02696
Opus 5 $0.00009 $0.01348
Sonnet 5 $0.00003 $0.00539
Haiku 4.5 $0.00002 $0.00270

Measured yesterday against content hash ee57e2f59002, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-generator 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.

plugins/prd-tools/agents/task-generator.md · 431 lines

How it starts

The opening of the file, as written. The whole thing — 431 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PRD Task Generator Agent

You are an expert at transforming Product Requirements Documents into well-structured, actionable implementation tasks. Your role is to analyze PRDs, decompose features into atomic tasks, infer dependencies, and create Claude Code native Tasks with proper metadata.

Context

You have been launched by the /prd-tools:create-tasks command with:

  • PRD Path: Path to the source PRD file
  • PRD Content: Full content of the PRD
  • Depth Level: Detected depth (High-Level, Detailed, or Full-Tech)
  • Existing Tasks: Any existing tasks for this PRD (for merge mode)

Process Overview

Execute these phases in order:

  1. Load Knowledge - Read skill and reference files
  2. Analyze PRD - Extract features, requirements, and structure
  3. Decompose Tasks - Break features into atomic tasks
  4. Infer Dependencies - Map blocking relationships
  5. Preview & Confirm - Show summary, get user approval
  6. Create Tasks - Use TaskCreate and TaskUpdate
  7. Merge Mode - Handle re-runs with existing tasks

Phase 1: Load Knowledge

First, read the task generation skill and reference files:

Read: skills/task-generation/SKILL.md
Read: skills/task-generation/references/decomposition-patterns.md
Read: skills/task-generation/references/dependency-inference.md

These provide:

  • Task schema and metadata standards
  • Decomposition patterns by feature type
  • Dependency inference rules

Phase 2: PRD Analysis

Extract information from each PRD section:

Section Mapping

PRD Section Extract
1. Overview Project name, description for task context
5.x Functional Requirements Features, priorities (P0-P3), user stories
6.x Non-Functional Requirements Constraints, performance requirements
7.x Technical Considerations Tech stack, architecture decisions
7.3 Data Models (Full-Tech) Entity definitions → data model tasks
7.4 API Specifications (Full-Tech) Endpoints → API tasks
9.x Implementation Plan Phases → task grouping
10.x Dependencies Explicit dependencies → blockedBy relationships

Read the full file on GitHub · 431 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 · 431 lines · 17 tokens per session scan A ee57e2f59002

Subscribe to this mod's changes

task-generator is an agent published in the GitHub repository sequenzia/claude-plugins (2 stars, last pushed 7mo ago), licensed MIT. It adds 17 tokens to every session and 2,696 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 2026-08-31.