recall
Reconstruct and narrate the current development context from contextual commits. Run at session start, when resuming work, or when switching branches. Produces a brief, conversational summary of where things stand.
80
72%
Does it follow best practices?
Impact
92%
1.35xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/recall/SKILL.mdQuality
Discovery
67%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description does a good job of explicitly stating both what the skill does and when to use it, which is its strongest aspect. However, it could benefit from more specific concrete actions (e.g., analyzing commit messages, summarizing file changes, identifying active features) and more natural trigger terms that users would actually say. The distinctiveness is moderate—it occupies a recognizable niche but could be confused with general git history or project status skills.
Suggestions
Add more natural trigger terms users would say, such as 'catch me up', 'what was I working on', 'project status', 'recent changes', or 'where did I leave off'.
List more specific concrete actions beyond 'reconstruct and narrate', e.g., 'analyzes recent commit messages, identifies active feature branches, summarizes file-level changes'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (development context from commits) and a key action (reconstruct and narrate), but doesn't list multiple concrete actions beyond producing a summary. 'Reconstruct and narrate' is somewhat specific but could be more detailed about what exactly is extracted or analyzed. | 2 / 3 |
Completeness | Clearly answers both what ('Reconstruct and narrate the current development context from contextual commits' and 'Produces a brief, conversational summary') and when ('Run at session start, when resuming work, or when switching branches'), with explicit trigger conditions. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'commits', 'branches', 'session start', 'resuming work', and 'switching branches', but misses common natural phrases users might say like 'what was I working on', 'catch me up', 'status', 'what changed', or 'context'. | 2 / 3 |
Distinctiveness Conflict Risk | Fairly specific to reconstructing development context from commits, but could overlap with git log/history skills or general project status skills. The 'session start' and 'resuming work' triggers help distinguish it, but 'development context' is somewhat broad. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted, highly actionable skill with clear workflows and executable commands for every scenario. Its main weakness is length — the comprehensive coverage of all scenarios, query modes, and example outputs makes it quite long for a single file, and some sections could be tightened or split out. The proactive usage section at the end is a valuable addition that rounds out the skill nicely.
Suggestions
Consider splitting Scope Query, Action+Scope Query, and Proactive Usage into a separate reference file, keeping SKILL.md focused on the default mode with clear links to the advanced query modes.
Trim the 'no contextual commits' example output — the paragraph suggesting adoption of the contextual-commit skill reads as filler and could be reduced to a single line.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly long (~300 lines) and includes some redundancy — e.g., the example outputs are extensive and partially repetitive across scenarios. However, most content is genuinely instructive and not explaining things Claude already knows. The bash snippets and example outputs are useful but could be tightened in places (e.g., the 'no contextual commits' fallback section includes a soft pitch paragraph that could be trimmed). | 2 / 3 |
Actionability | Excellent actionability throughout. Every scenario includes executable bash commands, concrete grep patterns, and detailed example outputs showing exactly what to produce. The argument detection logic, git commands, and extraction patterns are all copy-paste ready and specific. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced across all scenarios (detect branch state → gather raw material → extract action lines → synthesize output). Each scenario has explicit steps numbered and labeled. The argument detection at the top provides clear routing logic. The priority ordering for output synthesis is explicit. While there aren't traditional validation checkpoints, this is a read-only operation (querying git history) so destructive-operation feedback loops aren't needed. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear headers and scenario breakdowns, but it's entirely monolithic — everything lives in one file. The scope query, action+scope query, and proactive usage sections could reasonably be split into separate reference files, with SKILL.md providing the overview and default mode. The document is long enough that splitting would improve navigability. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- berserkdisruptors/contextual-commits
- Commit
8196cab
- 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.