Plan module

Skill Karthick-Ramachandran/persist-os/.agents/skills/plan-module

Local-first CLI that gives your repo durable, AI-ready memory - features, ADRs, modules, architecture, standards & tests and a deterministic doctor that validates health, evidence & drift.

Install

npx -y skills add Karthick-Ramachandran/persist-os --skill plan-module

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.5 KB, as published. Nobody here has run it

Skill: Plan Module

Purpose

Turn a module request into a complete delivery workflow before implementation starts.

Inputs

Module name or request.
Product goal or user need.
Existing architecture, feature, or module docs.
Known constraints or priority stage.

Required Reading

AGENTS.md
docs/00-product/PRD.md
docs/10-architecture/ARCHITECTURE.md
docs/20-security/SECURITY_MODEL.md
docs/50-quality/TESTING_STRATEGY.md
docs/60-engineering/ENGINEERING_STANDARDS.md
docs/ai/MODULE_DELIVERY_WORKFLOW.md
Relevant ADRs under docs/adrs/

Output Files

docs/40-features/F-###-<module>-module/ delivery docs (PRD, ACCEPTANCE, ARCHITECTURE_IMPACT, PLAN, TASKS, TEST_PLAN, REVIEW, COMPLETION_REPORT).
docs/30-modules/<module>/MODULE.md
docs/30-modules/<module>/TASKS.md
docs/30-modules/<module>/TEST_PLAN.md
docs/30-modules/<module>/DECISIONS.md

Process

Create a module brief: ownership, non-ownership, public interfaces, users, and use cases.
Define behavior and edge cases.
Write testable acceptance criteria.
Document architecture impact, dependency impact, config impact, template impact, and ADR needs.
Write the test plan from acceptance criteria, security invariants, and regression risk.
Break work into ordered tasks with status, scope, acceptance, tests, and do-not-do boundaries.
Mark implementation as blocked until the first task is selected.

Stop Conditions

Stop and request human decision if:

The module ownership or public interface is unclear.
The module conflicts with accepted ADRs or architecture docs.
The module conflicts with engineering standards.
The module requires runtime network, telemetry, cloud, MCP, AI API, auth, secrets, storage, or file write behavior changes without ADR or security review.
The user asks to implement before PRD, acceptance, architecture impact, test plan, and tasks exist.

Quality Bar

The module is planned as a mini product, not a file list.
PRD, acceptance, architecture impact, test plan, and tasks are all present.
Tasks are small enough to execute one at a time.
Module memory captures what future agents need to remember.
No implementation code is written by this skill.