CtrlK
BlogDocsLog inGet started
Tessl Logo

gitnexus-explorer

Serve an interactive codebase knowledge graph web UI.

56

Quality

66%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./optional-skills/research/gitnexus-explorer/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A strong, highly actionable skill body: every step is executable, the sequence is clear with a verify step, and the pitfalls capture real non-obvious failure modes. The main weaknesses are the inlined (and diverged) proxy script duplicating the unused `scripts/proxy.mjs` bundle file, and a few missing validation checkpoints around the patch/build and cleanup steps.

Suggestions

Replace the inline proxy code in Step 4 with a reference to the bundled script, e.g. "Copy `scripts/proxy.mjs` to `$GITNEXUS_DIR/proxy.mjs`" — the inline copy has already diverged from the (better) bundle version, which is exactly the drift risk progressive disclosure prevents.

Add a validation checkpoint after Step 2's build (e.g., confirm `npx vite build` exits 0 / `dist/index.html` exists) and before the cleanup's `rm -rf .claude/` and `gitnexus clean`, which are destructive and currently unvalidated.

Add a short failure-feedback note for step 5 (e.g., if the curl check returns 502, confirm `npx gitnexus serve` is running) to complete the error-recovery loop.

DimensionReasoningScore

Conciseness

The body is command-first with no explanations of concepts Claude already knows; every section earns its place (e.g., the cloudflared `--config /dev/null` pitfall is genuinely non-obvious). Not 5: the ~45-line inline proxy script and a few asides ("same origin, no CORS issues, no sudo, no nginx") could be trimmed or moved out of the main file.

4 / 5

Actionability

Every step is copy-paste executable: complete bash blocks, an exact patch shown as before/after TypeScript, the full proxy script, and a curl verification command. This matches the anchor for fully executable guidance covering the common case.

5 / 5

Workflow Clarity

A clear 7-step sequence with an explicit verification checkpoint ("Verify: `curl -s http://localhost:8888/api/repos` should return the indexed repo(s)") and a dedicated cleanup section. Not 5: there is no validation after the source patch/build step and no feedback loop if the tunnel or proxy fails; the destructive `rm -rf .claude/` and `gitnexus clean` cleanup steps run without a pre-check.

4 / 5

Progressive Disclosure

Section structure is good (When to Use, Prerequisites, Size Warning, Steps, Pitfalls), but the body inlines ~45 lines of proxy code as "Write this to a file" while `scripts/proxy.mjs` exists in the bundle and is never referenced — and the two copies have diverged (the inline version lacks the host-header fix, `.ttf`/`.map` MIME entries, and stricter SPA fallback present in the bundle file). This matches "content that should be separate is inline". Not 2: the rest of the document is well organized with no buried references.

3 / 5

Total

16

/

20

Passed

Description

53%Weight 40%Scale 1-5

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 states a clear, distinctive capability but is a single short sentence with no trigger guidance and thin action coverage. Adding a "Use when..." clause and a second concrete action (e.g., indexing a repo into a knowledge graph) would lift most dimensions.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user wants to visually explore a codebase's architecture or asks for a knowledge/dependency graph of a repo."

Include the indexing step in the what-statement ("Index a codebase into a knowledge graph and serve...") so specificity reflects the skill's full capability set.

Add natural synonym terms such as "dependency graph", "visualize code architecture", or "code exploration" to improve trigger-term coverage.

DimensionReasoningScore

Specificity

"Serve an interactive codebase knowledge graph web UI" names the domain and a single serving action, but omits the indexing, symbol/call-chain exploration, and tunneling capabilities the body actually delivers. Not 4: fewer than several specific concrete actions are listed.

3 / 5

Completeness

The "what" is clear (serve an interactive codebase knowledge graph web UI), but there is no "Use when..." clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

"codebase", "knowledge graph", and "web UI" are natural terms a user might say, but common variations like "dependency graph", "visualize architecture", or "code exploration" are missing. Not 4: keyword coverage is thin rather than good-with-a-few-gaps.

3 / 5

Distinctiveness Conflict Risk

"codebase knowledge graph web UI" carves a fairly distinct niche unlikely to fire for unrelated skills, with only minor overlap risk against generic codebase-inspection skills. Not 5: the description is short enough that it could collide with other visualization-oriented code skills.

4 / 5

Total

13

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository
NousResearch/hermes-agent
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.