here-be-git
Initialise a git repository with optional agent commit instructions and .gitignore. Use when users say "here be git", "init git", "initialise git", or otherwise indicate they want to set up version control in the current directory.
84
76%
Does it follow best practices?
Impact
100%
1.20xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/here-be-git/SKILL.mdQuality
Discovery
89%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 a well-crafted description that clearly defines its scope (git repo initialization) and provides explicit trigger guidance with both specific phrases and a general fallback. The main weakness is that the capabilities could be slightly more detailed about what exactly gets created or configured. Overall, it's a strong description that would perform well in skill selection.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (git repository) and some actions (initialise, optional agent commit instructions, .gitignore), but doesn't comprehensively list all concrete actions like creating specific files, configuring settings, or what 'agent commit instructions' entails. | 2 / 3 |
Completeness | Clearly answers both 'what' (initialise a git repository with optional agent commit instructions and .gitignore) and 'when' (explicit 'Use when' clause with specific trigger phrases and a general fallback condition). | 3 / 3 |
Trigger Term Quality | Includes natural trigger terms users would say: 'init git', 'initialise git', 'set up version control', and even a quirky but explicit phrase 'here be git'. Also covers the general case with 'otherwise indicate they want to set up version control'. | 3 / 3 |
Distinctiveness Conflict Risk | Very specific niche — git repository initialization — with distinct triggers like 'init git' and 'here be git' that are unlikely to conflict with other skills such as general git operations, committing, or branching. | 3 / 3 |
Total | 11 / 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 reasonably well-structured skill with a clear three-step interactive workflow. Its main weaknesses are the verbose, repetitive AGENTS.md content block that inflates the token count, and the lack of concrete executable examples for git commands and commit messages. The workflow logic is sound with good branching and edge case handling.
Suggestions
Trim the AGENTS.md commit instructions block significantly — the current version is extremely repetitive with the same point made 6+ different ways, which wastes tokens
Add concrete git command examples for commits, e.g., `git add AGENTS.md && git commit -m 'chore: add agent commit workflow instructions'`
Provide an actual .gitignore file example for at least one flavour rather than just listing pattern summaries with 'etc.'
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient but includes some unnecessary verbosity, particularly in the embedded AGENTS.md commit instructions which are extremely repetitive and emphatic. The .gitignore patterns section lists common knowledge that Claude already knows. However, the overall structure is reasonable. | 2 / 3 |
Actionability | The skill provides clear steps and specific content to add (the AGENTS.md block, .gitignore patterns), but lacks executable commands beyond `git init`. There are no concrete git commit command examples, no specific commit message examples, and the .gitignore patterns are listed as bullet summaries rather than actual file content. | 2 / 3 |
Workflow Clarity | The three-step workflow is clearly sequenced with explicit decision points (ask user, check if file exists, branch on yes/no). Each step has clear conditions and actions. The notes section handles edge cases like an already-initialized repo. The interactive prompts serve as natural checkpoints. | 3 / 3 |
Progressive Disclosure | The skill is a single file with no references to external files, which is acceptable for its size. However, the large embedded AGENTS.md markdown block makes the skill harder to scan and could arguably be referenced separately. The overall section structure is clear but the inline content block is bulky. | 2 / 3 |
Total | 9 / 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
- intellectronica/agent-skills
- Commit
9b0e00a
- 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.