write-exec-plan

Writes self-contained executable implementation plans with milestones, validation steps, and decision logs.

2|Updated May 6, 2026
One-click install
npx skills add https://github.com/bpcakes/jig-skills --skill write-exec-plan-bpcakes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: write-exec-plan
Source: https://github.com/bpcakes/jig-skills/tree/main/plugins/jig-exec-plans/skills/write-exec-plan
Command: npx skills add https://github.com/bpcakes/jig-skills --skill write-exec-plan-bpcakes

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Complex implementation tasks often fail because plans assume prior context, skip validation, or leave decisions undocumented. This Skill produces a fully self-contained ExecPlan that a novice agent or developer can follow end-to-end to deliver working, observable behavior. ## Core Features & Use Cases - Self-Contained Plan Authoring: Generates plans with all required context embedded, including file paths, commands, expected outputs, and plain-language definitions of jargon. - Living Document Structure: Enforces mandatory Progress, Surprises & Discoveries, Decision Log, and Outcomes & Retrospective sections that stay current as work proceeds. - Milestone and Validation Design: Structures work into independently verifiable milestones with behavior-based acceptance criteria, idempotent steps, and recovery guidance. - Use Case: When asked to plan a multi-stage feature like adding a background job queue, invoke this Skill to produce an ExecPlan covering every state-mutation path, exact test commands, and observable success criteria. ## Quick Start Ask the agent to write an ExecPlan for your feature, for example: write an executable implementation plan for adding rate limiting to the API with milestones and validation steps.

Frequently Asked Questions about write-exec-plan

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

FAQPage Schema
How do I write an executable implementation plan for a coding agent?

Invoke this Skill with a request for an ExecPlan describing your feature. It produces a self-contained document with purpose, milestones, concrete commands, validation steps, and decision logs that a stateless agent can follow without prior context.

What sections must an ExecPlan contain?

Every ExecPlan must include Progress, Surprises & Discoveries, Decision Log, and Outcomes & Retrospective sections, plus context, plan of work, concrete steps, validation, and idempotence guidance. These living sections must be updated as work proceeds.

When should I not trigger ExecPlan creation?

Do not create an ExecPlan for a complex task alone, a brief approach request, or when a plan already exists. The Skill activates only on a named invocation or an explicit request for a detailed implementation plan with milestones and validation.

Does writing an ExecPlan authorize the agent to implement it?

No. A plan-writing request does not authorize implementation, prototypes, commits, or deployment. Implementation only proceeds when separately included in the user's task scope.

How does the ExecPlan handle repository-specific plan standards?

The Skill reads PLANS.md or .agent/PLANS.md when present and follows those standards. If neither exists, it uses its built-in skeleton and notes that no repository-specific standard was found, without blocking or creating one.