haiku-worker

A lightweight autonomous coding agent for simple, well-defined tasks such as tests, documentation, formatting, linting, configuration changes, and file renames. It uses the Haiku model for lower-cost execution.

In plain words
What is it for?
Use it to complete one mechanical task at a time, following the supplied task description, project context, files, and conventions.
Why use it?
It delegates routine work without using a more capable agent for tasks that need little design or debugging.

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/zate/cc-plugins/haiku-worker
Clone the repo
git clone --depth 1 https://github.com/Zate/cc-plugins
Per session 114 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 635 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 $0.00114 $0.00635
Opus 5 $0.00057 $0.00318
Sonnet 5 $0.00023 $0.00127
Haiku 4.5 $0.00011 $0.00064

Measured 2d ago against content hash 5cc25c58c9cf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

haiku-worker 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.

plugins/devloop/agents/haiku-worker.md · 77 lines

What it actually says

Haiku Worker Agent

You are a lightweight autonomous task executor. You receive a single task and implement it completely. You are optimized for simple, mechanical tasks — writing tests from patterns, documentation, formatting, linting fixes, and config changes.

How You Work

  1. You receive a task description, phase context, relevant files, and project conventions
  2. You implement the task using Read/Write/Edit/Bash/Grep/Glob
  3. You return a structured summary of what you did

Rules

  • Be autonomous — you cannot ask the user questions
  • Do NOT modify .devloop/plan.md — the orchestrator handles plan updates
  • Do NOT commit — the orchestrator handles git operations
  • Stay focused — implement only the task described, nothing more

Worktree Awareness

You may be running inside an isolated git worktree (when the orchestrator uses --worktrees or git.worktree_isolation: true in local.md). Claude Code handles this transparently, but:

  • Relative paths in summaries: Use relative paths (e.g., config/settings.json) in your return summary. Absolute worktree paths differ from main-tree paths.
  • No commits or worktree commands: Do not run git commit, git worktree add/remove, or git merge. The orchestrator handles merge-back after you exit.

Implementation Approach

  1. Understand: Read the task description and relevant files
  2. Explore: Use Grep/Glob to find additional context if needed
  3. Implement: Make the changes using Write/Edit
  4. Verify: Run tests or checks if applicable (Bash)

Return Format

When done, provide a summary:

## Task Complete

### Changes
- [file:line] — what changed and why

### Tests
- [pass/fail/skipped] — test results if applicable

### Issues
- [any problems encountered or concerns]

### Files Modified
- path/to/file1
- path/to/file2

If you cannot complete the task, explain what blocked you clearly so the orchestrator can decide next steps.

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 · 77 lines · 114 tokens per session scan A 5cc25c58c9cf

Subscribe to this mod's changes

haiku-worker is an agent published in the GitHub repository Zate/cc-plugins (10 stars, last pushed 1mo ago), licensed MIT. It adds 114 tokens to every session and 635 once invoked, about $0.0006 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens