CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

skill-finish-branch

Wrap up a branch — run tests, create PR, merge or discard — use when implementation is done

Quality

69%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/skill-finish-branch/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%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 command-rich, actionable skill with genuine validation gates (tests, typed discard confirmation), undermined by structural noise: duplicated step numbering, an option-numbering mismatch between presentation and execution that risks performing the wrong action, an over-engineered review script, and redundant caution tables in an over-long monolithic file.

Suggestions

Fix the workflow coherence defects: remove the duplicate Step 4/Step 5 headings and align option numbering so "Present Options" and "Execute Choice" agree (currently 1=Create PR in presentation but Option 1=Merge Locally in execution), which could cause the wrong action to run.

Trim the Step 2 "quick review" down to the essential diff + reviewer invocation and move the defensive checks (binary/NUL/transport-marker/size limits, lifecycle hook wiring) into a bundled script under scripts/, cutting the bulk of the 80-line inline block.

Consolidate the overlapping "Red Flags", "Common Mistakes", and "Bottom Line" sections into one table, and replace placeholders like "npm test # or appropriate test command" and the missing changelog-prepend command with the concrete commands already shown in Step 1.

DimensionReasoningScore

Conciseness

The ~430-line body is noticeably padded: an 80-line over-engineered defensive review script (binary/NUL/transport-marker checks, lifecycle hook wiring) for a "quick review", duplicated Step 4 and Step 5 headings, three overlapping tables (Red Flags, Common Mistakes, Bottom Line) that repeat the same cautions, and niche octopus-integration sections. Not a 1 because nearly all content is commands rather than explanations of concepts Claude already knows.

2 / 5

Actionability

Most guidance is concrete and executable: test-suite detection script, git merge/push sequences, typed-confirmation discard flow, and a PR creation block with validated PR_TITLE/CHANGE_SUMMARY env guards. Minor gaps keep it from 5 — placeholders like "npm test # or appropriate test command", "# Prepend to CHANGELOG.md if it exists" with no actual command, and a hardcoded placeholder PR URL in the report.

4 / 5

Workflow Clarity

The pipeline sequence (tests → review → version → options → execute → cleanup) is present with real validation checkpoints (tests gate, review-clean check, typed "discard" confirmation, worktree cleanup rules), so it is not a 2. However, step numbers are duplicated (two "Step 4" and two "Step 5" headings) and the presented options contradict the execution sections — Step 5 presents "1. Create PR (Recommended), 2. Merge locally" while execution defines "Option 1: Merge Locally, Option 2: Create PR" — a mismatch that could execute the wrong user choice, which is more than the minor gaps of a 4.

3 / 5

Progressive Disclosure

The body is well-sectioned with a useful Quick Reference table, but it is a monolithic 430-line file with no bundle files (references/, scripts/, assets/ are absent), inlining an 80-line review orchestration script and per-option execution details that clearly belong in separate scripts or reference files. Structure exists, but content that should be separate is inline.

3 / 5

Total

12

/

20

Passed

Description

75%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 concise, third-person description that clearly names the domain and four concrete operations with an explicit use-when clause. Its main weaknesses are a generic trigger phrase and missing common synonyms (push, commit, finish) that would strengthen both completeness and trigger quality.

DimensionReasoningScore

Specificity

The description lists several concrete actions ("run tests, create PR, merge or discard") tied to the branch wrap-up domain, but it omits steps the skill actually performs (review, version bump, changelog, push, worktree cleanup), so coverage has minor gaps rather than being comprehensive.

4 / 5

Completeness

It answers both what ("run tests, create PR, merge or discard") and when ("use when implementation is done"), placing it above a 3; but the 'when' clause is generic and lacks concrete trigger phrases, so it does not reach the explicit 5 anchor.

4 / 5

Trigger Term Quality

Natural terms like "branch", "tests", "PR", "merge", "discard", and "implementation is done" are present, but common variations users would say — "push", "commit", "finish", "ship", "pull request" — are missing from the description itself (they appear only in the separate trigger field).

4 / 5

Distinctiveness Conflict Risk

"Wrap up a branch" with merge/PR/discard operations defines a fairly distinct niche with clear git-oriented triggers; the only overlap risk is with generic commit/git helper skills, which is minor.

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

nyldn/claude-octopus

Path

.claude/skills/skill-finish-branch/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.