codenavi
Your pathfinder for navigating unknown codebases. Investigates with precision, implements surgically, and never assumes — if it doesn't know, it says so. Maintains a .notebook/ knowledge base that grows across sessions, turning every discovery into lasting intelligence. Summons available skills, MCPs, and docs when the mission demands. Use when fixing bugs, implementing features, refactoring, investigating flows, or any development task in unfamiliar territory. Triggers on "fix this", "implement this", "how does this work", "investigate this flow", "help me with this code". Do NOT use for greenfield scaffolding, CI/CD, or infrastructure provisioning.
86
83%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
82%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 strong completeness with explicit 'Use when', 'Triggers on', and 'Do NOT use for' clauses, and good trigger term coverage with natural user phrases. However, it suffers from overly broad scope that could conflict with other development skills, and uses flowery/metaphorical language ('pathfinder', 'surgically', 'lasting intelligence') instead of concrete capability descriptions. The first-person-adjacent voice ('Your pathfinder') uses second person, which should be penalized per guidelines.
Suggestions
Replace metaphorical language ('pathfinder', 'surgically', 'lasting intelligence') with concrete actions like 'traces code flows, identifies dependencies, maps call chains'.
Narrow the scope or add more specific differentiators — 'any development task in unfamiliar territory' is too broad and will conflict with specialized coding skills. Consider specifying what makes this skill unique compared to general coding assistance.
Use third person voice ('Navigates unknown codebases...') instead of second person ('Your pathfinder...').
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (codebase navigation/development) and mentions some actions like investigating, implementing, and maintaining a .notebook/ knowledge base, but many phrases are vague and metaphorical ('pathfinder', 'surgically', 'lasting intelligence') rather than concrete specific actions. It doesn't list precise capabilities like 'traces function call chains' or 'maps dependency graphs'. | 2 / 3 |
Completeness | Clearly answers both 'what' (investigates codebases, implements changes, maintains .notebook/ knowledge base, summons skills/MCPs/docs) and 'when' with explicit 'Use when...' and 'Triggers on...' clauses, plus a 'Do NOT use for...' exclusion clause that further clarifies scope. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger phrases users would actually say: 'fix this', 'implement this', 'how does this work', 'investigate this flow', 'help me with this code'. Also includes broader terms like 'fixing bugs', 'implementing features', 'refactoring', 'investigating flows'. Good coverage of natural language variations. | 3 / 3 |
Distinctiveness Conflict Risk | The scope is quite broad — 'fixing bugs, implementing features, refactoring, investigating flows, or any development task' could overlap with many other coding-related skills. The 'unfamiliar territory' qualifier and the 'Do NOT use for' exclusions help somewhat, but triggers like 'fix this' and 'implement this' are very general and could conflict with more specialized coding skills. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-architected skill that provides a clear operational framework for navigating unknown codebases. Its greatest strengths are the explicit mission cycle with verification checkpoints, the Knowledge Verification Chain, and the detailed worked examples. The main weakness is moderate verbosity — some principles stated are things Claude would naturally do, and the Consistency Contract largely restates rules already covered in the Golden Rules and Mission Cycle sections.
Suggestions
Trim the Consistency Contract section — most items (e.g., #4 'never modify code outside scope', #9 'match existing code style') are already stated in the Golden Rules or Mission Cycle and don't need repetition.
Tighten the 'Adapting to Mission Scale' section to a compact table rather than prose blocks, since the scaling logic is straightforward.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is well-structured but verbose in places — the Golden Rules, Consistency Contract, and Mission Cycle explanations include some content Claude would naturally follow (e.g., 'match existing style', 'don't add features beyond what was asked'). The examples section is lengthy but valuable. Some sections like 'Adapting to Mission Scale' and the Consistency Contract could be tightened significantly. | 2 / 3 |
Actionability | The skill provides highly concrete, actionable guidance: specific plan templates with verification criteria, a clear Knowledge Verification Chain with numbered steps, explicit triggers for when to create/update/skip notebook entries, and three detailed worked examples showing the full cycle in practice. The mission cycle steps each have clear expected outputs. | 3 / 3 |
Workflow Clarity | The BRIEFING → RECON → PLAN → EXECUTE → VERIFY → DEBRIEF cycle is explicitly sequenced with clear validation checkpoints at each step. The Plan template requires verification criteria per step. The Verify step includes explicit fallback for when automated verification isn't possible. The Knowledge Verification Chain has a clear escalation path with a feedback loop (never skip to step 5 if 1-4 are available). | 3 / 3 |
Progressive Disclosure | The skill appropriately references external files for detailed content: 'references/coding-principles.md' for coding details and 'references/notebook-spec.md' for the notebook format specification. These are one-level-deep, clearly signaled references. The main file serves as a comprehensive overview without burying critical details in nested references. | 3 / 3 |
Total | 11 / 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
- tech-leads-club/agent-skills
- Commit
81e7e0d
- 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.