skills.rest
anderson-joyle
anderson-joyleCOMMUNITY
Explore all skills by this author →
GitHub

Executor

Execute runtime commands from typed JSON artifacts and write auditable result files.

2|Updated Apr 26, 2026
Search Tags:#worktree#execution#command runner#audit logs#probe gate#patch apply#typed artifacts
One-click install
npx skills add https://github.com/anderson-joyle/claude-a-team --skill executor-anderson-joyle
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Executor
Source: https://github.com/anderson-joyle/claude-a-team/tree/main/skills/executor
Command: npx skills add https://github.com/anderson-joyle/claude-a-team --skill executor-anderson-joyle

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill ensures AI-generated work is executed safely and reproducibly by applying typed artifacts, running only explicitly requested commands, and producing auditable execution results with no guessing.

Core Features & Use Cases

  • Probe-mode command execution: Runs probe_commands from probe-output.json, captures stdout/stderr, and writes probe-results.json without applying any file changes.
  • Implementation-mode change application: Applies planned_changes and runs execution_handoff.test_commands only after the Probe Gate passes, then writes execution-results.json.
  • Strict input validation and fail-loud behavior: Refuses to run when required inputs are missing or invalid, records failures with reasons, and stops rather than continuing after planned-change failure.

Quick Start

Run the Executor after the Engineer stage to produce probe-results.json in probe mode or execution-results.json in implementation mode, using the session worktree and the required upstream artifacts.

Frequently Asked Questions about Executor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run probe commands and capture stdout/stderr without applying file changes?▼

Probe command execution runs specified commands from probe-output.json, captures stdout and stderr logs, and writes probe-results.json without applying any file changes to ensure safe testing.

How do I apply planned changes and run test commands deterministically?▼

Implementation-mode change application applies planned changes from engineer-output.json and runs execution_handoff test commands only after Probe Gate passes, writing execution-results.json with auditable outputs.

What happens when required input artifacts are missing or invalid for command execution?▼

Strict input validation enforces fail-loud behavior by refusing to run when required inputs are missing or invalid, recording failures with specific reasons, and stopping rather than continuing after a planned-change failure.

Do I need a worktree to run deterministic file operations and patch applications?▼

Worktree-only file operations are required, enforcing that all patch contexts and file applications occur strictly within the session worktree to validate paths and maintain auditable execution results.

When do I need typed JSON artifacts for reproducible runtime command execution?▼

Typed JSON artifacts are needed when running deterministic runtime commands in the A-TEAM pipeline, requiring probe-output.json or engineer-output.json plus probe-gate-output.json in implementation mode to produce auditable results.

What's the best way to audit command execution results and patch applications in a CI pipeline?▼

Auditable execution results are produced by running only explicitly requested commands, validating patch contexts, capturing stdout/stderr logs, and writing execution-results.json or probe-results.json with correct envelope fields.

Related Skills

Explore More Software Engineering Skills→
keshav55keshav55

email-agent

Manage Gmail via AtrisOS to read, send, and archive emails.

Community
Intermediate
Bhavya-DhootBhavya-Dhoot

aras-results-analyst

Interpret ARAS results.json into a narrative and markdown table.

Community
Advanced
2lab-ai2lab-ai

setup-ai-workflow

Configure an AI work-execution system with gate checks, commit/PR conventions, worktree isolation, and deployment runbook.

Official
Advanced
masermediagroup-stackmasermediagroup-stack
1

shell

Execute user-provided shell commands and return exit status and output.

Community
Basic