CtrlK
BlogDocsLog inGet started
Tessl Logo

ci-artifact-testing

Use when validating an upstream pull request or dependency fix against VS Code tests using GitHub Actions CI build artifacts instead of building locally. Covers finding compatible binaries, verifying PR and merge-commit provenance, selecting a candidate runtime or build, comparing unchanged regression tests, and restoring the baseline.

75

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 strong, disciplined workflow document: executable commands with placeholders, a five-phase process with explicit validation and stop conditions, and thorough cleanup/reporting guidance. The only weaknesses are mild — a few sentences could be trimmed and some detail could move to reference files for a leaner overview.

Suggestions

Move the GitHub Actions discovery/download command blocks (sections 1–2) into a references file (e.g., references/github-actions.md), keeping one short example inline in SKILL.md, so the overview stays lean.

Tighten the stacked prohibitions in sections 1 and 3 by merging related 'Do not...' clauses (e.g., combine the substitution/bypass and retry/assertion-weakening rules into single compact lines).

Trim explanatory rationale sentences like 'Busy Repositories can push the desired artifact off that page' to their operative instruction ('List artifacts for the selected run, not the repository-wide latest-artifacts page').

DimensionReasoningScore

Conciseness

The body is efficient and assumes Claude's competence (no explanation of what GitHub Actions or CI is; every section carries non-obvious operational knowledge such as merge-commit provenance and artifact layout). Not a 5 because a few rationale-heavy sentences ('Busy Repositories can push the desired artifact off that page', several stacked 'Do not...' clauses) could be tightened; not a 3 because there is no over-explanation of known concepts.

4 / 5

Actionability

The gh CLI examples are copy-paste ready with variables defined and placeholders explicitly flagged ('replace the illustrative values'): 'gh pr view "$pr" --repo "$repo" --json url,state,headRefOid,baseRefOid', 'gh run download "$run_id" --repo "$repo" --name "$artifact_name" --dir "$artifact_dir"', plus the commit-parents provenance check. Non-code guidance is equally concrete (e.g., 'Restore permissions only on identified executables'). Not a 4 because the command path covers the common GitHub Actions case completely end to end.

5 / 5

Workflow Clarity

A clear five-step sequence (identify, verify provenance/download, select, compare, restore/report) with explicit validation checkpoints and error-recovery guidance: 'Stop or qualify the result when source provenance cannot be established', 'Record its actual failure; a startup error or fixture mismatch is not the original regression', 'Repeat lifecycle- or timing-sensitive scenarios at least twice', 'Check the final diff/status against the starting state'. Not a 4 because checkpoints are explicit at every phase, including the restore phase.

5 / 5

Progressive Disclosure

No bundle files exist, so everything is inline; sections are well organized and sibling-skill links ('[integration-test](../integration-tests/SKILL.md)', '[azure-pipelines](../azure-pipelines/SKILL.md)') are clearly signaled and one level deep. Not a 5 because at ~95 lines the body exceeds a single-screen overview — the GitHub Actions discovery command blocks and the artifact-layout/permissions detail would fit naturally in a references file; not a 3 because navigation is easy and nothing is buried or nested.

4 / 5

Total

18

/

20

Passed

Description

100%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.

A model description: it names a specific niche, explicitly states when to trigger with natural user phrasing, and enumerates the concrete capabilities of the workflow. It is dense yet unpadded.

DimensionReasoningScore

Specificity

The description lists five concrete actions ('finding compatible binaries, verifying PR and merge-commit provenance, selecting a candidate runtime or build, comparing unchanged regression tests, and restoring the baseline') in a named domain (VS Code tests via GitHub Actions CI build artifacts), matching the comprehensive-coverage anchor. Not a 4 because coverage spans the entire workflow rather than having minor gaps.

5 / 5

Completeness

Both questions are explicitly answered: 'Use when validating an upstream pull request or dependency fix against VS Code tests using GitHub Actions CI build artifacts instead of building locally' gives the when, and 'Covers finding compatible binaries, ... restoring the baseline' gives the what with concrete trigger phrases. Not a 4 because neither half is vague or implicit.

5 / 5

Trigger Term Quality

Natural user phrases are comprehensively covered: 'upstream pull request', 'dependency fix', 'VS Code tests', 'GitHub Actions CI build artifacts', 'building locally', with synonyms present (pull request/PR, artifacts/binaries). Not a 4 because the common variations a user would actually say are all present; file extensions are not applicable to this scenario.

5 / 5

Distinctiveness Conflict Risk

The trigger is a clear niche (validating upstream fixes against VS Code tests using CI artifacts rather than local builds), distinct from generic build/test or artifact-download skills. Not a 4 because the combination of 'VS Code tests' and 'CI build artifacts' leaves minimal overlap risk with closely related skills.

5 / 5

Total

20

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 5 suspicious

Warning

Total

15

/

16

Passed

Repository
microsoft/vscode
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.