CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pre-publish

Validate a package before publishing to npm. Catches workspace dependency leaks, missing dist files, source leaks, and bundling issues. TRIGGER when: publishing to npm, modifying package.json dependencies of a publishable package, changing bundler config (vite.config, tsup.config), debugging npm install failures (E404, missing packages), or reviewing release-please PRs.

71

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

77%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced validation checklist with executable commands, expected outputs, and rich incident-driven error recovery. Its weaknesses are redundancy — the private-dependency rule is repeated across five-plus sections — and a monolithic single-file layout with no progressive disclosure, where the post-mortem archive and CI details could be split into reference files.

Suggestions

State the 'private @moltnet/* bundled deps go in devDependencies; published @themoltnet/* stay in dependencies' rule once (e.g., in checklist item 2 or 4) and have the legreffier, pi-extension, and SDK-pattern sections reference it instead of restating it, cutting substantial repeated text.

Move the 'Common mistakes and how they happen' post-mortems and the 'Integration with CI' section into a references/ file (e.g., references/incidents.md), keeping SKILL.md as a lean checklist that links to it one level deep.

Trim narrative detail in the incident sections to the failing signal, why it was missed, and the prevention step — the September 2026 majors entry in particular interleaves history that can be compressed.

DimensionReasoningScore

Conciseness

The body avoids explaining concepts Claude already knows and is dense with repo-specific rules, but the core 'private @moltnet/* deps belong in devDependencies' rule is restated in at least five places (checklist items 2 and 4, the legreffier and pi-extension incidents, the SDK pattern, and the reminders), and the post-mortem narratives run ~90 lines. Anchor 3 fits: mostly efficient but could be tightened; the redundancy exceeds the 'minor instances' of anchor 4.

3 / 5

Actionability

Every checklist item ships copy-paste-ready commands with expected outcomes ('rg -n '@moltnet/' <package>/package.json', 'pnpm exec nx run <project>:check:pack', 'pnpm pack --dry-run --json | jq', a full temp-dir tarball install script) plus a concrete repository JSON snippet. Not 4 because the commands are executable as written with clearly marked placeholders and the common cases are covered.

5 / 5

Workflow Clarity

A nine-step sequenced checklist with explicit validation at each step (expected rg match counts, check:pack as a gate that enumerates what it validates, tarball inspection, clean-consumer smoke test), error-recovery guidance via the 'Common mistakes' post-mortems, and CI integration showing where checks gate the merge and publish. Matches the clear-sequence-with-explicit-validation-and-checklists anchor.

5 / 5

Progressive Disclosure

Everything lives in one ~320-line SKILL.md with no references/, scripts/, or assets/ files; the ~90-line incident archive and the CI integration section are inlined content that would fit a references file, and only one external pointer exists (the ../rendered-pack-3252bc08/SKILL.md link, clearly signaled). Good section headers keep it above anchor 2, but the structure is a single file where a split is warranted.

3 / 5

Total

16

/

20

Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An excellent description: concrete capabilities, an explicit 'TRIGGER when' clause with natural file names and error codes, and a clearly bounded npm-publish niche. The only weakness is that two of its triggers (npm install failure debugging, bundler config changes) overlap with general npm/bundler troubleshooting skills.

DimensionReasoningScore

Specificity

Names the concrete action ('Validate a package before publishing to npm') and enumerates specific failure classes it catches: 'workspace dependency leaks, missing dist files, source leaks, and bundling issues' — comprehensive coverage of the niche. Not 4 because the enumerated failure classes cover the domain's common cases rather than leaving minor gaps.

5 / 5

Completeness

Explicitly answers both: what ('Validate a package before publishing to npm. Catches workspace dependency leaks...') and when ('TRIGGER when: publishing to npm, modifying package.json dependencies... or reviewing release-please PRs') with concrete trigger phrases. Not 4 because the when-clause is fully explicit and enumerated rather than merely present.

5 / 5

Trigger Term Quality

Covers natural terms users would actually say: 'publishing to npm', 'package.json dependencies', 'vite.config, tsup.config', 'E404, missing packages', 'release-please PRs' — including file names and error codes, matching the comprehensive-with-extensions anchor. Not 4 because no commonly used synonym or trigger phrasing for this domain is obviously missing.

5 / 5

Distinctiveness Conflict Risk

The publish/release framing ('publishing to npm', 'release-please PRs', 'publishable package') gives it a clear niche, but 'debugging npm install failures (E404, missing packages)' and 'changing bundler config (vite.config, tsup.config)' are triggers a general npm-troubleshooting or bundler-config skill could also claim. Fits 'mostly distinct; minor overlap risk' better than the minimal-conflict anchor 5.

4 / 5

Total

19

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

getlarge/themoltnet

Path

.agents/skills/pre-publish/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.