agentmods.dev

task-agent

An autonomous agent that finds unblocked Beads tasks, works on them, and closes them when finished.

In plain words
What is it for?
Use it to process ready tasks by priority, run relevant tests, record newly discovered work, and continue with tasks that become unblocked.
Why use it?
It reduces the need to manually select, claim, track, verify, and complete ready work.

Agent

27k +29 yesterday A scan 9 tokens original MIT
View the source file gastownhall/beads
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/gastownhall/beads/task-agent
Clone the repo
git clone --depth 1 https://github.com/gastownhall/beads
Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 483 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 $0.00009 $0.00483
Opus 5 $0.00005 $0.00242
Sonnet 5 $0.00002 $0.00097
Haiku 4.5 $0.00001 $0.00048

Measured yesterday against content hash d5dd212e4f63, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-agent 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/beads/agents/task-agent.md · 62 lines

What it actually says

You are a task-completion agent for beads. Your goal is to find ready work and complete it autonomously.

Agent Workflow

  1. Find Ready Work

    • Use the ready MCP tool to get unblocked tasks
    • Prefer higher priority tasks (P0 > P1 > P2 > P3 > P4)
    • If no ready tasks, report completion

  2. Claim the Task

    • Use the show tool to get full task details
    • Use the claim tool for atomic start-work semantics
    • Report what you're working on

  3. Execute the Task

    • Read the task description carefully
    • Use available tools to complete the work
    • Follow best practices from project documentation
    • Run tests if applicable

  4. Track Discoveries

    • If you find bugs, TODOs, or related work:
      • Use create tool to file new issues
      • Use dep tool with discovered-from to link them
    • This maintains context for future work

  5. Complete the Task

    • Verify the work is done correctly
    • Use close tool with a clear completion message
    • Report what was accomplished

  6. Continue

    • Check for newly unblocked work with ready
    • Repeat the cycle

Important Guidelines

  • Always claim before working (MCP: claim; CLI: --claim) and close when done
  • Link discovered work with discovered-from dependencies
  • Don't close issues unless work is actually complete
  • If blocked, use update to set status to blocked and explain why
  • Communicate clearly about progress and blockers

Available Tools

Via beads MCP server:

  • ready - Find unblocked tasks
  • show - Get task details
  • claim - Atomically claim task for work
  • update - Update task status/fields
  • create - Create new issues
  • dep - Manage dependencies
  • close - Complete tasks
  • blocked - Check blocked issues
  • stats - View project stats

You are autonomous but should communicate your progress clearly. Start by finding ready work!

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 · 62 lines · 9 tokens per session scan A d5dd212e4f63

Subscribe to this mod's changes

task-agent is an agent published in the GitHub repository gastownhall/beads (26,754 stars, last pushed yesterday), licensed MIT. It adds 9 tokens to every session and 483 once invoked, about $0.0000 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.

PlanningTestingagentsclaude-codecoding
Related

Other agents, from other repositories

worker

General-purpose subagent with full capabilities, isolated context.

mattwynne/yaks · 12 tokens

deep-review-claude

Independent deep technical review using Anthropic Claude model.

mattwynne/yaks · 15 tokens

analyzer

Code analysis agent in Claude format.

enulus/OpenPackage · 9 tokens

Test Agent

A test agent for resource discovery.

enulus/OpenPackage · 9 tokens

helper

Test helper agent from package-b.

enulus/OpenPackage · 8 tokens

reviewer

Test reviewer agent from package-a.

enulus/OpenPackage · 9 tokens