CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

comparison-base

Internal reusable workflow for resolving the correct Git comparison base for branch reviews and change analysis. Finds the PR target or canonical Fluid Framework upstream, selects the newest shared target-history commit, and detects substantial divergence.

Quality

80%

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

Fix and improve this skill with Tessl

tessl review fix ./.github/skills/comparison-base/SKILL.md

The canonical home for this skill is comparison-base in microsoft/FluidFramework

SKILL.md
Quality
Evals
Security

Quality

Content

81%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 procedural skill body: ordered resolution steps, executable git/gh commands, explicit validation checkpoints, and well-handled ambiguity via user prompts. Minor improvements are possible in tightening the dense Inputs prose and supplying commands for the shallow-repo deepening and metadata-recording steps.

DimensionReasoningScore

Conciseness

The body is procedural with no explanations of concepts Claude already knows, and the gh/git gotchas ('Do not rely on bare gh pr view…') earn their tokens. Not 5: the Inputs paragraph is a dense wall of prose and a few sentences ('Keep these values distinct…') could be tightened into a list.

4 / 5

Actionability

Provides concrete, executable commands for the core steps — 'git fetch --no-tags "$TARGET_SOURCE" "+refs/heads/$TARGET_BRANCH:refs/comparison/target"', 'gh pr list --repo <candidate-base-repository> --state open --head …', 'git merge-base --all', 'git rev-list --count'. Not 5: two steps lack commands — deepening shallow histories and recording the comparison commit's short SHA/date/subject are described but not executable.

4 / 5

Workflow Clarity

Clear three-step sequence with explicit validation ('Verify $BASE_COMMIT is non-empty and is an ancestor of both $TARGET_REF and $REVIEW_REF'), ordered PR-resolution sources with stop conditions, and feedback loops for error recovery (nonzero merge-base status, empty base in shallow repos, multiple best ancestors). Risky actions are gated — merging the target branch 'never perform it without explicit approval'.

5 / 5

Progressive Disclosure

A single well-organized file with clear section headers (Inputs, Steps 1–3, Outputs, Edge Cases) and no nested or buried references; nothing clearly belongs in a separate file. Not 5: at ~110 lines it exceeds the under-50-line single-file exception, and the Edge Cases section partially restates guidance already embedded in the steps.

4 / 5

Total

17

/

20

Passed

Description

72%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 specific, well-scoped description with strong third-person action coverage and a clear niche. Its main weakness is trigger-term coverage: natural synonyms like 'merge base' and 'base branch' are absent, and the use-when guidance is contextual rather than explicit.

Suggestions

Add natural trigger synonyms such as 'merge base', 'base branch', and 'diff base' so the description matches how users actually phrase the need.

Convert the contextual 'for branch reviews and change analysis' into an explicit trigger clause, e.g. 'Use when a workflow needs to diff a branch against its merge target.'

DimensionReasoningScore

Specificity

Lists multiple concrete third-person actions — 'resolving the correct Git comparison base', 'Finds the PR target or canonical Fluid Framework upstream', 'selects the newest shared target-history commit', 'detects substantial divergence' — with comprehensive coverage of the workflow. Not 4: the action list covers the full capability rather than leaving minor gaps.

5 / 5

Completeness

Clearly answers 'what' (find PR target/upstream, select newest shared commit, detect divergence) and includes a 'when' via 'for branch reviews and change analysis'. Not 5: the 'when' is contextual rather than an explicit 'Use when…' trigger clause with concrete trigger phrases.

4 / 5

Trigger Term Quality

Relevant keywords are present ('Git', 'comparison base', 'branch reviews', 'change analysis', 'PR target', 'divergence') but common natural variations are missing — most notably 'merge base', 'base branch', and 'diff'. Matches anchor 3 rather than 4 because the canonical term a user would most naturally say ('merge base') never appears.

3 / 5

Distinctiveness Conflict Risk

A clear niche is carved by 'Git comparison base', 'PR target', and 'canonical Fluid Framework upstream', making confusion unlikely. Not 5: the trigger context 'branch reviews and change analysis' is generic enough to minorly overlap with general code-review skills.

4 / 5

Total

16

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

microsoft/FluidFramework

Path

.github/skills/comparison-base/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.