he-prune-branches
Review and prune stale branches safely. Use when branch cleanup needs evidence, protected-branch caution, PR awareness, and non-destructive recommendations.
59
68%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./Plugins/harness-engineering/fixtures/budget-archive/2026-04-21/deferred-store/skills/team_automation/he-prune-branches/SKILL.mdQuality
Discovery
75%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description has good structure with an explicit 'Use when' clause and a distinctive niche, but the specificity of concrete actions is moderate and the trigger terms could better cover natural user language variations. The 'when' clause focuses more on qualities of the approach (evidence, caution, non-destructive) rather than user-facing trigger scenarios.
Suggestions
Add more concrete actions like 'list branches by last commit age, check merge status, identify branches with closed PRs, generate safe delete commands'.
Include more natural trigger terms users would say, such as 'git branches', 'delete old branches', 'merged branches', 'remote branch cleanup', or 'branch housekeeping'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (branch cleanup) and some actions ('review', 'prune stale branches'), but lacks concrete specifics like listing branches by age, checking merge status, generating delete commands, or identifying merged vs unmerged branches. | 2 / 3 |
Completeness | Clearly answers both 'what' (review and prune stale branches safely) and 'when' (when branch cleanup needs evidence, protected-branch caution, PR awareness, and non-destructive recommendations) with an explicit 'Use when' clause. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'branches', 'branch cleanup', 'PR', and 'protected-branch', but misses common natural user phrases like 'git branches', 'delete old branches', 'stale branches', 'merged branches', 'remote branches', or 'housekeeping'. | 2 / 3 |
Distinctiveness Conflict Risk | The description carves out a clear niche around stale branch pruning with specific qualifiers (evidence-based, protected-branch caution, PR awareness, non-destructive), making it unlikely to conflict with general git skills or other repository management tools. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a competent skill with strong workflow clarity—the multi-step procedure is well-sequenced with explicit confirmation gates, validation checkpoints, and error recovery guidance. Its main weaknesses are incomplete actionability (referenced script not in bundle, vague inputs/outputs, no concrete output example) and some unnecessary boilerplate sections that reduce conciseness without adding value.
Suggestions
Provide a concrete example of the expected output format (the 'schema_version: 1' structure) so Claude knows exactly what to produce.
Include the referenced script 'clean-gone' in the bundle or inline its logic, since the skill depends on it for the discovery step.
Replace the vague 'Inputs' section with specific parameter descriptions or remove it entirely if inputs are self-evident from context.
Remove the 'Full Context' icon asset references—they add no instructional value and waste tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some sections that could be tightened—'Inputs' is vague and unhelpful, 'Full Context' listing icon assets adds no value, and some constraint bullets feel boilerplate rather than skill-specific. However, it avoids explaining concepts Claude already knows. | 2 / 3 |
Actionability | The procedure references a concrete script path and provides executable bash commands for worktree/branch deletion, which is good. However, the script itself is not provided in the bundle, the 'Inputs' section is vague ('Request, artifacts, repo context'), and the 'Outputs' section describes a schema without showing a concrete example. Key details like how to detect the active branch or check PR status are missing. | 2 / 3 |
Workflow Clarity | The procedure is clearly sequenced with explicit validation checkpoints: discovery → __NONE__ exit path → present candidates → single confirmation gate → worktree removal before branch deletion → report outcomes. The Validation section adds a fail-fast gate and covers edge cases like active branch and root worktree protection. The Anti-Patterns section reinforces correct ordering. | 3 / 3 |
Progressive Disclosure | The skill is well-organized with clear section headers and logical flow. However, no bundle files are provided despite referencing a script path, and the 'Full Context' section only links to icon assets rather than useful reference material. The skill is somewhat monolithic—the Validation, Constraints, and Anti-Patterns sections could potentially be split out for a cleaner overview, though the total length is manageable. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |
- Repository
- jscraik/Agent-Skills
- Commit
4c78f98
- 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.