Identify and fix stale task states in your project tracker.

Install

mkdir -p .claude/skills/cleanbeads && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12560" && unzip -o skill.zip -d .claude/skills/cleanbeads && rm skill.zip

Installs to .claude/skills/cleanbeads

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Audit the beads tracker for stale or incorrectly-marked work — in_progress claims no agent is actually working, beads stuck in_progress that should be blocked, beads stuck blocked whose dependencies have closed, deferred beads whose defer date has passed, and ready-queue beads with an assignee set but never claimed. Cross-references active processes, uncommitted edits in the shared checkout, and the agent-mail coord thread to tell the difference between "actively running" and "stale." Asks per-bead before applying any state change. Default staleness threshold is 48 hours; pass an arg like `/cleanbeads 24h` or `/cleanbeads 7d` to override.
646 charsno explicit “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Advanced

Key capabilities

  • Identify stale `in_progress` beads.
  • Detect `in_progress` beads that should be `blocked` due to dependencies.
  • Find `blocked` beads that should be `open` because blockers are closed.
  • Surface `deferred` beads whose defer date has passed.
  • Locate `open` beads with an assignee set but not claimed.
  • Propose corrections for bead states and apply them with `bd update`.

How it works

The skill gathers live-work signals, classifies `in_progress` beads, checks blocker graphs for `blocked` beads, verifies defer dates for `deferred` beads, and identifies owner-set but unclaimed `open` beads, then presents findings and applies user-approved changes.

Inputs & outputs

You give it
An optional staleness threshold (e.g., `24h`, `7d`).
You get back
A summary of bead audit findings, proposed state changes, and updated bead states in `.beads/beads.jsonl`.

When to use cleanbeads

  • Cleaning up stale tasks
  • Resetting stuck progress claims
  • Validating project tracker data

About cleanbeads

Performs an audit on task states to identify forgotten or stale claims. It proposes corrections to ensure tracking accuracy.

Audit the beads tracker for stale or incorrectly-marked work — in_progress claims no agent is actually working, beads stuck in_progress that should be blocked, beads stuck blocked whose dependencies have closed, deferred beads whose defer date has passed, and ready-queue beads with an assignee set b

When not to use it

  • When running in the middle of an active `/nextbeads` or `/analyzesat` run.
  • When the user wants to create or close beads.
  • When the user wants to auto-claim ready beads.

Limitations

  • It does not create or close beads.
  • It does not auto-claim ready beads.
  • It does not edit bead descriptions, acceptance criteria, or labels.

How it compares

This skill systematically audits bead states, cross-referencing live processes and git status to distinguish active work from stale claims, which is more accurate and automated than manual tracking.

Compared to similar skills

cleanbeads side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
cleanbeads (this skill)02moReviewAdvanced
ito-fix02moNo flagsBeginner
triage03moNo flagsBeginner
archon99moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

ito-fix

withakay

Start a fix-oriented Ito change proposal with fix-biased intake and schema recommendations.

00

triage

junghoon-son

ED-style severity assessment for any incoming task, bug, request, or alert. Sorts work by acuity using a Claw-General-flavored ESI (Emergency Severity Index). Use BEFORE deciding what to work on next, or whenever a backlog needs ordering.

00

archon

coleam00

Interactive Archon integration for knowledge base and project management via REST API. On first use, asks for Archon host URL. Use when searching documentation, managing projects/tasks, or querying indexed knowledge. Provides RAG-powered semantic search, website crawling, document upload, hierarchical project/task management, and document versioning. Always try Archon first for external documentation and knowledge retrieval before using other sources.

9140

linear

lobehub

Linear issue management guide. Use when working with Linear issues, creating issues, updating status, or adding comments. Triggers on Linear issue references (LOBE-xxx), issue tracking, or project management tasks. Requires Linear MCP tools to be available.

10117

zapier-workflows

davila7

Manage and trigger pre-built Zapier workflows and MCP tool orchestration. Use when user mentions workflows, Zaps, automations, daily digest, research, search, lead tracking, expenses, or asks to "run" any process. Also handles Perplexity-based research and Google Sheets data tracking.

11101

attio-skill-generator

kesslerio

Generate use-case-specific Attio workflow skills from templates. Use when creating new skills for lead qualification, deal management, customer onboarding, or custom Attio workflows.

7100

Search skills

Search the agent skills registry