Jujutsu (jj) — the Git-compatible version control system. Activate ONLY when a .jj/ directory is present in the project or when jj/jujutsu is explicitly mentioned. Do NOT activate for plain git repos without .jj/. Use for any VCS operations in jj-managed projects: commit, push, pull, branch, bookmark, rebase, squash, merge, diff, log, status, working copy, change ID, revset, fileset, template, configuration, workspaces.
99
100%
Does it follow best practices?
Impact
100%
1.28xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Quality
Discovery
100%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 is an excellent skill description that clearly identifies the tool, provides comprehensive trigger terms covering both common VCS operations and jj-specific concepts, and explicitly defines activation boundaries to avoid conflicts with Git skills. The negative trigger condition ('Do NOT activate for plain git repos without .jj/') is a particularly strong element that minimizes conflict risk.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and concepts: commit, push, pull, branch, bookmark, rebase, squash, merge, diff, log, status, working copy, change ID, revset, fileset, template, configuration, workspaces. Also clearly identifies the tool (Jujutsu/jj) and its relationship to Git. | 3 / 3 |
Completeness | Clearly answers both 'what' (VCS operations in jj-managed projects with a comprehensive list) and 'when' (explicit activation criteria: .jj/ directory present OR jj/jujutsu explicitly mentioned, with a clear negative condition for plain git repos). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: includes both 'jj' and 'jujutsu', '.jj/ directory', and a comprehensive list of VCS operation keywords (commit, push, pull, branch, rebase, etc.) plus jj-specific terms like 'bookmark', 'revset', 'fileset', 'change ID', and 'workspaces'. | 3 / 3 |
Distinctiveness Conflict Risk | Exceptionally distinct — explicitly differentiates itself from plain git by stating 'Do NOT activate for plain git repos without .jj/' and specifies the unique activation trigger (.jj/ directory). This makes it very unlikely to conflict with a standard Git skill. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is an exceptionally well-crafted skill file. It efficiently conveys jj's unique mental model and non-obvious behaviors, provides fully actionable commands with agent-specific safety rules, and structures content with clear progressive disclosure to detailed reference files. The common mistakes table and agent rules demonstrate deep understanding of failure modes in automated contexts.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section earns its place. The mental model section conveys non-obvious concepts (working copy is a commit, change IDs vs commit IDs, bookmarks ≠ branches) that Claude wouldn't inherently know about jj. No padding or explanation of basic VCS concepts. The tables are dense and scannable. | 3 / 3 |
Actionability | Provides fully executable bash commands and TOML config throughout. The core workflow, pushing changes, recovery, and essential commands table are all copy-paste ready with concrete flags and arguments. Agent-specific rules are precise and operational (e.g., always use `-m`, never use interactive commands). | 3 / 3 |
Workflow Clarity | The core workflow is clearly sequenced (describe → code → new → repeat) with explicit validation steps ('jj st && jj diff' after changes, 'verify after mutations' rule). The agent rules section provides a clear feedback loop (verify with `jj st` after destructive ops). Recovery workflow is explicit with undo/op log/op restore progression. The non-linear work section includes a decision checkpoint ('decide if it depends on the current chain'). | 3 / 3 |
Progressive Disclosure | Excellent two-tier structure: the Topics table provides clear navigation to deep-dive files, inline '→ Deep dive' links connect specific sections to relevant detail files, and the Reference Index at the bottom organizes all supporting materials by category. References are one level deep and clearly signaled throughout. | 3 / 3 |
Total | 12 / 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 |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
40067f1
Table of Contents
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.