CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

managing-git-workflow

Git and pull-request workflow for HASH, including conditional Linear conventions and merge-queue diagnostics. Use when naming a branch, opening or reviewing a pull request, interpreting its checks, or diagnosing a merge-queue ejection.

69

Quality

87%

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

86%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 well-crafted instruction skill: it is dense with project-specific, non-obvious facts, provides exact formats, examples, and commands, and defers policy to clearly signaled source-of-truth documents instead of duplicating them. The only weakness is that a few merge-queue verification and diagnosis steps name endpoints and refs without fully assembled commands, leaving minor executable gaps.

Suggestions

Provide assembled commands for merge-queue verification (e.g., a complete `gh api` call for the issue timeline and a `gh run list` invocation filtered to `merge_group` runs on the `gh-readonly-queue/...` ref), mirroring the fully written-out commands in the Reviewing section.

Close the reviewing workflow with an explicit next step after findings are identified (e.g., how blocking findings versus optional suggestions should be communicated and resolved), turning the implicit checkpoint into an explicit one.

DimensionReasoningScore

Conciseness

The body explains nothing Claude already knows about Git or pull requests; every sentence carries a HASH-specific, non-obvious rule (e.g., "The PR's `auto_merge` field remains `null` after a successful enqueue, so it is not evidence either way" and the clippy/cargo-hack divergence selected through `GITHUB_EVENT_NAME`). Lean and efficient; every token earns its place, matching anchor 5 rather than anchor 4's 'minor instances of over-explanation'.

5 / 5

Actionability

Concrete, executable guidance dominates: copy-paste commands ("gh pr view <PR_NUMBER> --comments", the full "gh api" call with headers), exact branch format with worked examples ("ln/fe-1437-hash-monorepo-import"), an exact PR-title pattern, and exact commands ("cargo clippy --all-features", "cargo hack --optional-deps --feature-powerset clippy"). It falls just short of anchor 5 because the merge-queue verification steps name endpoints and refs ("GET /repos/hashintel/hash/issues/<NUMBER>/timeline", the "gh-readonly-queue/<base>/pr-<NUMBER>-<sha>" ref) without assembled commands.

4 / 5

Workflow Clarity

Each workflow has a clear ordered sequence with concrete discriminating checks: contributing (branch name → PR title → self-contained description), reviewing (gather record and diff → establish baseline contract → separate blocking findings from suggestions), and ejection diagnosis ("Compare the failing step against the same step on the identical head in the PR-level run"). It sits at anchor 4 rather than 5 because checkpoints are implicit in the prose rather than explicit validate/recover gates — e.g., the reviewing flow never states what to do after findings are identified.

4 / 5

Progressive Disclosure

The body is compact and well-sectioned with no content inlined that belongs in a separate file, and its external references are one level deep and clearly signaled as sources of truth ("Treat [`.github/CONTRIBUTING.md`]... as the source of truth for contribution policy", plus the organization-wide PR template link). No bundle files exist and none are needed; navigation is easy, matching anchor 5. It does not exhibit anchor 4's 'minor organization gaps'.

5 / 5

Total

18

/

20

Passed

Description

83%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 strong description: it states a clear 'what', an explicit and concrete 'Use when...' trigger list, and distinctive HASH/Linear/merge-queue specifics. The main improvement space is trigger-term synonym coverage (e.g., 'PR', 'code review') and making the 'what' clause enumerate capability verbs rather than only naming the domain.

DimensionReasoningScore

Specificity

The description lists several concrete actions — "naming a branch, opening or reviewing a pull request, interpreting its checks, or diagnosing a merge-queue ejection" — plus scoped sub-areas ("conditional Linear conventions and merge-queue diagnostics"). It is not a 5 because the 'what' clause names the domain and its components rather than enumerating capability actions; the concrete verbs all live in the trigger clause.

4 / 5

Completeness

Both questions are answered explicitly: the 'what' ("Git and pull-request workflow for HASH, including conditional Linear conventions and merge-queue diagnostics") and an explicit "Use when..." clause with four concrete trigger activities. This matches the anchor-5 pattern of a clear 'what' plus concrete trigger phrases, and exceeds anchor 4, whose 'when' is less explicit or specific.

5 / 5

Trigger Term Quality

Natural user phrases are present: "naming a branch", "opening or reviewing a pull request", "interpreting its checks", "diagnosing a merge-queue ejection". A few common variations are missing — the "PR" abbreviation, "code review", "commit", and the unhyphenated "merge queue" — which fits anchor 4 (good coverage, a few natural terms missing) rather than anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

"merge-queue ejection", "Linear conventions", and the HASH scoping carve out a clear niche with distinct triggers. However, "naming a branch" and "opening... a pull request" are generic triggers that a general git-workflow skill would also claim, so minor overlap risk with closely related skills keeps this at anchor 4 rather than 5.

4 / 5

Total

17

/

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: 1 suspicious

Warning

Total

15

/

16

Passed

Repository

hashintel/hash

Path

.agents/skills/managing-git-workflow/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.