agentsclimarketplace

Routine

Skill menot-you/n-soul/skills/routine

Memory-safe engineering-peer plugin for Claude Code + Codex: 11 skills, 6 bounded agents, 4 hooks. Claims only what it can prove; closes every turn with one next action.

Install
npx -y skills add menot-you/n-soul --skill routine

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

2 things to look at

  • 15 days oldThe repository was created 15 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 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.

What its author says it does

Copied from the file, not written here

Execute bounded engineering work through current-state checks, minimal changes, focused proof, and observed verification.

SKILL.md

1.8 KB, 333 tokens by cl100k_base, as published. Nobody here has run it

Routine

Use for a clear, reversible engineering task whose result and scope are already locked. Keep one active result, one proof path, and one next action visible.

Inputs to lock

  • Exact result and acceptance proof.
  • Allowed files and explicit exclusions.
  • Current branch, working tree, and overlapping work.
  • Risk, authority, rollback, and stop conditions.

Method

  1. Read the current target files and repository guidance.
  2. Trace changed symbols to direct callers and affected processes. If no code graph is available, label the manual trace as limited.
  3. Observe the failing state or missing behavior before editing when the result can be tested.
  4. Choose the smallest clean change that removes the cause. Do not hide a failure behind retries, fallbacks, or unrelated refactors.
  5. Match local style and remove only dead code created by this change.
  6. Add or update focused tests that fail before the change and pass after it.
  7. Run the narrow check first, then every broader gate required by the touched surface.
  8. Review the diff for scope drift, security expansion, generated artifacts, and rollback.
  9. Confirm no file outside the allowed scope changed.
  10. Report the strongest proven state, observed verification, rollback status, and any remaining gap.

After two missed changes, stop. Switch to the debug skill and obtain new proof before another patch.

Completion

A routine is complete only when the requested state is observed, required checks are green, scope is clean, and rollback remains available. Local green is not proof of merge, release, or live reachability. Do not close with status theater.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.