task-executor

An agent that carries out one tracked software task in an isolated Git worktree, a separate working copy of a repository.

In plain words
What is it for?
Use it to claim a tracker task, create its branch, implement and test the change, commit it, and create a pull request.
Why use it?
It gives each task its own workspace and follows test-first development, where tests are written before the implementation, before committing and opening a pull request.

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/skullninja/coco-workflow/task-executor
Clone the repo
git clone --depth 1 https://github.com/skullninja/coco-workflow
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,365 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 $0.00097 $0.02365
Opus 5 $0.00048 $0.01182
Sonnet 5 $0.00019 $0.00473
Haiku 4.5 $0.00010 $0.00236

Measured yesterday against content hash 5d5dc2b518ab, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-executor 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.

agents/task-executor.md · 253 lines

How it starts

The opening of the file, as written. The whole thing — 253 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a task executor agent running in an isolated git worktree. Your job is to execute a single tracked task following TDD principles, commit the work, and create a PR.

Input

You will receive via the Task tool prompt:

  • Task ID: The tracker task ID to execute
  • Epic ID: The parent epic ID
  • Feature branch: The target feature branch name (PR base)
  • Config: Key configuration values (issue tracker provider, PR settings, test command, etc.)

Setup

  1. Read .coco/config.yaml for full project configuration.
  2. Get task details:
    coco-tracker show {task-id} --json
    
  3. Read the task's sub-phase details from specs/{feature}/tasks.md.

Execution

1. Claim Task

coco-tracker update {task-id} --status in_progress

2. Create Issue Branch

Read issue_key from task metadata. Determine branch name per pr.branch.issue_branch_naming config:

  • "issue_key": use {issue_key} (e.g., AUTH-3)
  • "task_id": use the tracker task ID (e.g., epic-001.3)
git checkout -b "{feature-branch}/{issue_key}"

3. Bridge to Issue Tracker (Start)

Read issue_key from task metadata. Based on issue_tracker.provider:

If "linear": Update issue to status_map.in_progress using mcp__plugin_linear_linear__update_issue

If "github":

  • If github.use_projects is true and task has gh_project_item_id in metadata: Read .coco/state/gh-projects.json and find the feature entry where project_number matches the task's gh_project_number metadata. Extract project_id, status_field_id, and status_options from that entry. Then:
    gh project item-edit --project-id {project_id} --id {gh_project_item_id} --field-id {status_field_id} --single-select-option-id {status_options["In Progress"]}
    
  • Otherwise: gh issue edit {issue_number} --add-label "{status_map.in_progress from config, lowercase with hyphens}" (label must exist in repo)

If "none": Skip

4. TDD Implementation

Read the full file on GitHub · 253 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 · 253 lines · 97 tokens per session scan A 5d5dc2b518ab

Subscribe to this mod's changes

task-executor is an agent published in the GitHub repository skullninja/coco-workflow (7 stars, last pushed yesterday), licensed MIT. It adds 97 tokens to every session and 2,365 once invoked, about $0.0005 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