implement-plan

Validate plan, task, and context artifacts and write a structured execution report.

9|1|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/conn-castle/agent-layer --skill implement-plan-conn-castle
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: implement-plan
Source: https://github.com/conn-castle/agent-layer/tree/main/internal/templates/skills/implement-plan
Command: npx skills add https://github.com/conn-castle/agent-layer --skill implement-plan-conn-castle

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates the disciplined execution of plan-driven work by validating and coordinating plan, task, and context artifacts, and by generating a reproducible run report.

Core Features & Use Cases

  • Artifact discovery under .agent-layer/tmp to locate the latest plan and task pair (and optional context).
  • Phase-driven execution with governance decisions: proceed, revise, escalate, or rewrite-because-out-of-scope.
  • Deterministic reporting: writes .agent-layer/tmp/implement-plan.<run-id>.report.md for audit and repeatability.
  • Multi-agent coordination: coordinates Context scout, Execution gatekeeper, Implementer, and Verifier for safe execution.

Quick Start

Invoke the implement-plan workflow to execute the latest plan-task-context trio and generate a run report.

Frequently Asked Questions about implement-plan

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

FAQPage Schema
How do I automate workflow orchestration for multi-agent task execution?

Workflow orchestration for multi-agent task execution is automated by validating plan, task, and context artifacts, coordinating execution across specialized agents, and writing a deterministic run report for audit and repeatability.

What is phase-driven execution with gating decisions in repository automation?

Phase-driven execution with gating decisions is a workflow control mechanism that evaluates artifacts at each stage to proceed, revise, escalate, or rewrite out-of-scope tasks, ensuring disciplined and safe plan execution.

How do I generate deterministic run reports for plan-driven work?

Deterministic run reports for plan-driven work are generated by discovering matching plan-task-context sets, executing gated phases, and writing the final structured output to a uniquely identified report file for repeatability.

Do I need a context file to execute plan and task artifacts?

A context file is optional for executing plan and task artifacts, but the workflow requires discovering the latest matching plan-task pair under the designated temporary repository directory to proceed.

Can I coordinate multiple agents like a scout and gatekeeper for task execution?

Coordinating multiple agents for task execution is supported by orchestrating a Context scout, Execution gatekeeper, Implementer, and Verifier to validate artifacts, enforce governance decisions, and ensure safe execution.

What are the limitations of artifact discovery for plan execution?

Artifact discovery for plan execution is limited to locating the latest matching plan-task-context trio under the specific repository temporary directory and enforces strict artifact naming rules to produce valid deterministic outputs.