claude-execute-detailed

Executes a detailed implementation plan with verification, acceptance checks, and PR publication.

2|1|Updated Jun 25, 2026
One-click install
npx skills add https://github.com/Consiliency/agent-harness --skill claude-execute-detailed-consiliency
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-execute-detailed
Source: https://github.com/Consiliency/agent-harness/tree/main/skills-src/claude/claude-execute-detailed
Command: npx skills add https://github.com/Consiliency/agent-harness --skill claude-execute-detailed-consiliency

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It turns a bounded, detailed implementation plan into verified, committed code changes without scope creep, unsafe git operations, or unverified closeouts. ## Core Features & Use Cases - Bounded plan execution: Reads a detailed plan artifact and applies only the changes declared in its ## Changes section, stopping on any out-of-scope edit. - Verification and acceptance reduction: Runs every command in the plan's ## Verification section and checks each ## Acceptance criteria item, reporting pass/fail per item. - Safe git workflow and PR publication: Establishes a clean worktree off a fresh base ref, never commits to protected branches, and publishes a draft or ready PR via the runtime publication primitive. - Use Case: After generating a detailed implementation plan, hand it to this Skill to implement the changes in an isolated worktree, run the test suite, and open a pull request with a full verification report. ## Quick Start Execute the detailed plan at the given artifact path, run its verification commands, and open a draft PR with the results.

Frequently Asked Questions about claude-execute-detailed

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

FAQPage Schema
How do I execute a detailed implementation plan with Claude Code?▼

Provide the detailed plan artifact path, or let the skill resolve the latest valid handoff from claude-plan-detailed. It validates the plan's Changes, Verification, and Acceptance criteria sections, then applies edits, runs verification, and reports results.

What happens if plan verification fails during execution?▼

The skill attempts one local diagnosis and repair, then stops with the diagnostic 'verification failed' if the command still fails or is out of scope. Work is preserved on a safe branch and the failure is recorded in the manifest lifecycle and reflection.

Does this skill commit directly to the main branch?▼

No. It never commits to main or protected branches. It creates a dedicated worktree and branch off a fresh base ref, then publishes a draft or ready PR from that branch using the runtime publication primitive.

When does it open a draft PR instead of a ready PR?▼

A draft PR is opened when dependencies remain or verification was partial or skipped. A ready PR requires complete verification; skipped or partial verification never produces a ready PR.

What are the limitations of bounded plan execution?▼

It cannot expand beyond the plan's declared Changes section, cannot read protected files unless the plan allowlists them, and stops on non-plan output rather than silently widening scope. Multi-phase work with lanes belongs to a phase runner instead.