j.dev-done

Chains commit and mirror-sync commands to sync framework changes into .claude/ and .agents/ mirrors.

Updated Aug 3, 2026
One-click install
npx skills add https://github.com/samwelmunga/jenga-npm --skill j-dev-done-samwelmunga
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: j.dev-done
Source: https://github.com/samwelmunga/jenga-npm/tree/main/.agents/skills/j-dev-done
Command: npx skills add https://github.com/samwelmunga/jenga-npm --skill j-dev-done-samwelmunga

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? After committing framework-level changes (skills, agents, hooks, scripts, templates), the .claude/ and .agents/ mirror directories can sit stale until someone remembers to run a separate sync command. This Skill removes that gap by chaining the commit and mirror-sync steps into one invocation, while safely halting when there is nothing to commit. ## Core Features & Use Cases - Chained Commit and Sync: Invokes /commit with a scope-id (epic, story, or task id such as E42_S04_T01), then automatically runs /self-sync to update the .claude/ and .agents/ mirrors. - Deterministic Halt Detection: A bash classifier script inspects the captured /commit output for the exact message "No implementation to commit." and stops the chain before /self-sync runs. - Collision-Safe Alias: Provides a guaranteed-unshadowed invocation path when a host tool's built-in command shadows the bare /dev-done name. - Use Case: After implementing a root-level framework change, run the skill with the task scope-id to commit the work and immediately propagate it to both mirror directories in one step. ## Quick Start Ask the agent to run dev-done with your task scope-id, for example "dev-done E42_S04_T01", to commit the current work and sync the mirrors.

Frequently Asked Questions about j.dev-done

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

FAQPage Schema
How do I commit and sync mirror directories in one step?▼

Invoke the skill with a scope-id such as an epic, story, or task id (e.g. E42_S04_T01). It runs /commit with that scope-id, then automatically runs /self-sync to update the .claude/ and .agents/ mirrors.

What happens if there is nothing to commit?▼

A bash classifier script checks the captured /commit output for the exact message "No implementation to commit." If found, it exits with code 1, prints that message, and the chain stops without running /self-sync.

Why use j-dev-done instead of the bare dev-done command?▼

Some host tools have built-in commands that shadow the bare /dev-done name because skill resolution is a literal directory-name match. The j-dev-done directory provides a guaranteed-unshadowed alias with identical behavior.

Does the sync still run if commit reports drift or doc-sync findings?▼

Yes. Drift and doc-sync findings from /commit are informational, not blocking. The chain only halts on the exact "No implementation to commit." message; otherwise /self-sync always runs.

Can I edit the j-dev-done SKILL.md file directly?▼

No. The file is generated by scripts/generate-j-alias.sh from skills/dev-done/SKILL.md. Hand edits are overwritten; re-run the generator to pick up source changes.