arn-code-assess
This skill should be used when the user says "arness code assess", "arn-code-assess", "assess codebase", "technical review", "codebase assessment", "find improvements", "what should I improve", "tech debt review", "tech debt audit", "pattern compliance check", "codebase health check", "assess the project", "improvement plan", "review my codebase", "what needs fixing", "code quality check", "audit my code", "run an assessment", or wants a comprehensive technical assessment of the codebase against stored patterns followed by prioritized improvement execution through the full Arness pipeline.
56
64%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/arn-code/skills/arn-code-assess/SKILL.mdQuality
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 excels at trigger term coverage with an extensive list of natural phrases users might say, and it clearly addresses both 'what' and 'when'. However, the actual capability description is somewhat vague—it mentions 'stored patterns' and 'full Arness pipeline' without explaining what concrete actions are performed or what outputs are produced. The heavy reliance on trigger phrase enumeration compensates for the lack of specificity in the capability description.
Suggestions
Add specific concrete actions the skill performs, e.g., 'Scans codebase for anti-patterns, checks compliance against stored best-practice patterns, generates a prioritized list of improvements with severity ratings, and produces an actionable remediation plan.'
Reduce the trigger phrase list to the most distinctive terms and instead use that space to describe concrete outputs and capabilities of the assessment pipeline.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description mentions 'comprehensive technical assessment of the codebase against stored patterns' and 'prioritized improvement execution through the full Arness pipeline', which names the domain and some actions but doesn't list multiple specific concrete actions (e.g., what exactly does the assessment check, what does the pipeline produce). | 2 / 3 |
Completeness | The description explicitly answers 'when' with a comprehensive list of trigger phrases and 'what' by describing it as a comprehensive technical assessment against stored patterns with prioritized improvement execution. The 'Use when' equivalent is clearly present via 'This skill should be used when...'. | 3 / 3 |
Trigger Term Quality | The description includes an extensive list of natural trigger phrases users would actually say, such as 'assess codebase', 'tech debt review', 'code quality check', 'what needs fixing', 'review my codebase', and many more variations covering both formal and informal phrasings. | 3 / 3 |
Distinctiveness Conflict Risk | While the Arness-specific triggers ('arness code assess', 'arn-code-assess') are distinctive, many generic triggers like 'code quality check', 'what needs fixing', 'audit my code' could easily overlap with other code review or linting skills. The mix creates moderate conflict risk. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
47%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill excels at workflow clarity with thorough decision gates, resumability, conflict detection, and error handling — it's a well-designed orchestration pipeline. However, it is far too verbose for its purpose as a sequencer; the content could be cut by 40-50% without losing actionable information. The repeated progress bar displays, exhaustive option listings, and detailed prose for each gate inflate token cost significantly while the actual executable guidance (code, commands) is minimal.
Suggestions
Drastically reduce verbosity: collapse the 7 gate descriptions into a compact reference table with columns for gate ID, trigger condition, question, and options — the current prose format repeats the same pattern 7 times with unnecessary padding.
Remove the repeated ASCII progress bar displays from every step — define the format once and say 'show progress indicator with current stage highlighted' in each step.
Inline the key content from referenced files (assessment-protocol.md, orchestration-flow.md) or provide them as bundle files, since the skill currently defers critical logic to files that aren't available.
Add concrete examples of agent invocation syntax (actual Tool call format) rather than describing them in prose — this would significantly improve actionability.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines. It over-explains every gate, step, and edge case in exhaustive detail. Much of this content (how to present options, how to track state, how to display progress bars) is procedural orchestration that Claude can infer from a more compact description. The repeated progress bar displays alone consume significant tokens for minimal value. | 1 / 3 |
Actionability | The skill provides concrete step sequences, gate definitions, and specific tool invocations (Skill tool, Agent tool references). However, it lacks executable code examples — the 'code' blocks are just ASCII progress bars. Key details like the assessment protocol and orchestration flow are deferred to reference files that aren't provided, making the skill incomplete on its own. | 2 / 3 |
Workflow Clarity | The workflow is exceptionally well-sequenced with 13 numbered steps, 7 explicit decision gates in a clear table, resumability detection with an artifact-to-state mapping table, conflict detection between specs, and explicit error recovery paths (retry/skip/abort). Validation checkpoints are present at testing (G6) and conflict detection (G5). | 3 / 3 |
Progressive Disclosure | The skill references external files (assessment-protocol.md, orchestration-flow.md, ensure-config.md) which is good progressive disclosure in principle, but no bundle files are provided so we can't verify they exist or are well-structured. The main SKILL.md itself is monolithic — much of the detailed gate logic, error handling, and state detection tables could be split into reference files to keep the overview lean. | 2 / 3 |
Total | 8 / 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 | |
- Repository
- AppsVortex/arness
- Commit
b9084b6
- 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.