agentsclimarketplace

Propose catalogue pack

Skill eugenelim/agent-ready-repo/.agents/skills/propose-catalogue-pack

Use to justify and scaffold a NEW pack area for this catalogue — test that it is additive and fits the catalogue's declared coverage model plus the four charter principles, then scaffold the pack shell and emit an RFC with a per-primitive inventory, or reject it as non-additive. Triggers on "should we add a pack for <area>", "propose a new pack", "justify a <vendor/domain> pack". Do NOT use to ingest units (use assimilate-primitive or assimilate-repo).From its SKILL.md

Install
npx -y skills add eugenelim/agent-ready-repo --skill propose-catalogue-pack

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

SKILL.md

3.5 KB, as published. Nobody here has run it

Skill: propose-catalogue-pack

Stand up a new pack the right way: prove it earns its place, scaffold the shell to convention, and route the decision through an RFC — or reject it. Justification-first; the scaffold is the reward for clearing the bar, not the starting point.

Output rendering

Table — When presenting several items that share the same fields, render a Markdown table. Cap at ~5 columns; beyond that, switch to a per-item detail list. Right-align numeric columns. Rationale / narrative — Use short ## headings and 2–3 sentence paragraphs. Don't force narrative into a table.

Procedure

  1. Test additivity + fit against the local charter. Read this catalogue's docs/CHARTER.md coverage model (SDLC here; a different model in a re-purposed fork) and the four charter principles — universal · substantive not duplicative · a habit not a tool · used often enough to stick. A proposed area that duplicates an existing pack, or fails a principle, is a reject with the failing principle named.

    Tech-stack accelerator packs are a distinct routing path — their specificity is the point, so they are exempt from Principle 1 (Universal) and must instead clear Principles 2–4 plus three extra gates: (1) a named maintainer, (2) a stated maturity scope (validated / contract-complete / experimental), and (3) an archiving/deprecation path. An accelerator pack that cannot satisfy all three extra gates is a reject. If the proposal is tech-stack-specific (a CI/CD platform, IaC tool, SaaS integration), route it through this path explicitly — do not fail it on Principle 1.

  2. Diagnose the boundary. What primitives would the pack carry, what does it depend on (core? another pack?), what's explicitly out of scope. If a heavy-knowledge dependency on another folder path in the same source is discovered, surface it as a blocker — a pack that can't stand on its own isn't additive.

  3. Scaffold the shell (only on a pass) — pack.toml, .claude-plugin/plugin.json, README.md, empty .apm/. Route all writes through agentbundle.safety.write_jailed. See references/pack-shell.md.

  4. Emit an RFC with the per-primitive inventory + verdicts (the pack's candidate skills/agents/hooks, each assimilate/reject). When the inventory comes from a survey, consume assimilate-repo's output rather than re-inventorying.

  5. Prepare elicitation, don't flood. Where fit, naming, or scope is a judgment, present what you found + options + a recommendation.

Never do

  • Write under this repo's packages/agentbundle/** or packs/credential-brokers/**.
  • Scaffold a pack that hasn't cleared the additivity + four-principles bar — reject non-additive areas explicitly rather than shipping a shell.
  • Write outside agentbundle.safety.write_jailed.

Depends on core + governance-extras. Repo-scope; not in any default profile.

What ships with it: 1 file

1.7 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most test skills give

Counted across 1,201 of the 2,096 authors here whose files we hold, read 2026-09-06

  • Write a failing test before writing codein 43 of 1201, across 36 files
  • Run the full test suitein 36 of 1201, across 35 files
  • Test only one variable per experimentin 34 of 1201, across 17 files
  • Read product marketing context before asking questionsin 34 of 1201, across 14 files
  • Mock external dependenciesin 34 of 1201, across 30 files
  • Define primary, secondary, and guardrail metricsin 33 of 1201, across 16 files
  • Pre-determine sample size before startingin 31 of 1201, across 14 files
  • Test behavior rather than implementationin 31 of 1201, across 29 files
  • Formulate a hypothesis before designing a testin 30 of 1201, across 13 files
  • Document every test hypothesis, variant, and resultin 29 of 1201, across 11 files
  • Use descriptive test function namesin 25 of 1201, across 21 files
  • Commit to the methodology without stopping earlyin 24 of 1201, across 8 files

Said here and by no other author read

  • test additivity against the local charter
  • verify pack against the four charter principles
  • validate accelerator packs against three extra gates
  • diagnose pack boundaries and dependencies
  • scaffold pack shell only after passing tests
  • route all file writes through safety tools

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.

Keep looking

Skills are one crate of 613,872. 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.