task-planner

An agent for managing individual task files stored in .claude/tasks/, a folder where project TODOs persist between sessions.

In plain words
What is it for?
Use it to create, list, update, archive, and report on tasks, including warnings for overdue or inactive work.
Why use it?
It keeps tasks, statuses, priorities, dates, and stale work visible without relying on memory or one temporary checklist.

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/osisdie/claude-code-channels/task-planner
Clone the repo
git clone --depth 1 https://github.com/osisdie/claude-code-channels

Made for: Claude Code.

Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 820 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 $0.00029 $0.00820
Opus 5 $0.00015 $0.00410
Sonnet 5 $0.00006 $0.00164
Haiku 4.5 $0.00003 $0.00082

Measured 2d ago against content hash 784777cd92d3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-planner 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-planner.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Task Planner Agent

Manage persistent TODO tasks as atomic markdown files in .claude/tasks/. Each task is a separate file that survives context compression.

Operations

Execute the operation specified in the prompt. Always use UTC+8 timestamps.

list

  1. Glob .claude/tasks/*.md (skip .archive/ subdirectory)
  2. Read the YAML frontmatter of each file
  3. Return a markdown table sorted by: priority (high → medium → low), then status (in_progress → pending → done)
  4. Flag any task with status: in_progress and updated_at older than 24 hours as [stale]
  5. Flag any task with status: pending and created_at older than 3 days as [stale]
  6. If no task files exist, return: "No active tasks."
  7. After the table, add a summary line: "N pending, N in_progress, N done. M stale."
  8. If 3+ tasks are pending or any task is stale, append: "⚠ Recommend running a task review."

Output format:

| ID  | Title                   | Status      | Priority | Updated      |
|-----|-------------------------|-------------|----------|--------------|
| 001 | Fix SVG font fallback   | in_progress | high     | 2h ago       |
| 002 | Add dark-bg watermark   | pending     | high     | 1d ago [stale] |

create

  1. Glob .claude/tasks/*.md to find the highest numeric prefix
  2. Compute next ID = max + 1 (start at 001 if empty)
  3. Generate slug from title (lowercase, hyphens, max 40 chars)
  4. Create directory .claude/tasks/ if it doesn't exist: mkdir -p .claude/tasks
  5. Write the task file with this template:
---
id: {N}
title: "{title}"
status: pending
priority: {priority}
created_at: "{ISO 8601 timestamp}"
updated_at: "{ISO 8601 timestamp}"
---

## Description
{description}

## Acceptance Criteria
{criteria as checkbox list, or "TBD" if not specified}
  1. Return: "Created task {NNN}: {title}"

update {ID}

  1. Glob .claude/tasks/{ID}-*.md to find the file
  2. Read it, modify the specified frontmatter fields
  3. Always update updated_at to current timestamp
  4. Optionally append to ## Notes section if notes provided
  5. Write the file back
  6. Return: "Updated task {ID}: {changes}"

Read the full file on GitHub · 102 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 · 102 lines · 29 tokens per session scan A 784777cd92d3

Subscribe to this mod's changes

task-planner is an agent published in the GitHub repository osisdie/claude-code-channels (5 stars, last pushed 5mo ago), licensed MIT. It adds 29 tokens to every session and 820 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.