CtrlK
BlogDocsLog inGet started
Tessl Logo

skill-iter-tune

Iterative skill tuning via execute-evaluate-improve feedback loop. Uses ccw cli Claude to execute skill, Gemini to evaluate quality, and Agent to apply improvements. Iterates until quality threshold or max iterations. Triggers on "skill iter tune", "iterative skill tuning", "tune skill".

62

Quality

73%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/skill-iter-tune/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

47%

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

This skill is a well-structured orchestration document with excellent workflow clarity and clear phase sequencing, but it suffers significantly from verbosity — it reads more like an architecture design document than a concise skill instruction. The extensive pseudocode, ASCII diagrams, and inline implementation details (termination logic, TodoWrite patterns, coordinator checklists) bloat the content far beyond what's needed for an orchestrator that delegates to phase files. The actionability is moderate since the actual executable details are deferred to phase files not included in the bundle.

Suggestions

Reduce the SKILL.md to a lean orchestrator overview (~80-100 lines max) by moving termination logic, TodoWrite patterns, error handling tables, and the coordinator checklist into their respective phase files or a shared utilities reference.

Replace the illustrative JavaScript pseudocode with actual executable commands or tool invocations that Claude can directly use, or clearly mark them as the phase files' responsibility and remove them from the orchestrator.

Consolidate the three separate flow diagrams (Architecture Overview, Chain Mode Extension, Data Flow) into a single concise diagram — they largely repeat the same information.

Move the Interactive Preference Collection code block and the detailed TodoWrite Pattern into the phase files (e.g., 01-setup.md) since they are implementation details, not orchestration-level concerns.

DimensionReasoningScore

Conciseness

Extremely verbose at ~300+ lines. Contains extensive architecture diagrams, JavaScript pseudocode for orchestration logic, TodoWrite patterns, and coordinator checklists that could be dramatically condensed. Much of this (termination logic, error handling tables, data flow diagrams) is implementation detail that bloats the orchestrator document. The ASCII art diagrams, while visually appealing, consume significant tokens for information that could be stated in 1-2 sentences.

1 / 3

Actionability

Provides structured pseudocode for the iteration loop, termination logic, and preference collection, but none of it is truly executable — it's illustrative JavaScript that references undefined functions (AskUserQuestion, TaskCreate, TaskUpdate) and tools (ccw cli) without complete, copy-paste-ready commands. The actual execution details are deferred to phase files that aren't provided, making the skill more of a design document than actionable instructions.

2 / 3

Workflow Clarity

The multi-step workflow is exceptionally well-sequenced with clear phase ordering (1→2→3→4→5), explicit termination conditions (threshold, max iterations, convergence, error limit), error recovery strategies per phase with rollback mechanisms, and validation checkpoints (check termination before Phase 4, snapshot before execute). The iteration loop structure with feedback is clearly defined.

3 / 3

Progressive Disclosure

The skill correctly references phase documents (phases/01-setup.md through phases/05-report.md) with a clear table mapping phases to documents and purposes, following a one-level-deep pattern. However, since no bundle files are provided, we cannot verify these references exist. More critically, the SKILL.md itself contains too much inline detail (full pseudocode for termination, TodoWrite patterns, error handling) that should live in the phase files, undermining the progressive disclosure principle it espouses.

2 / 3

Total

8

/

12

Passed

Description

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 a strong description that clearly communicates a specific workflow (execute-evaluate-improve loop), names the tools involved (ccw cli Claude, Gemini, Agent), and provides explicit trigger phrases. The description is concise yet comprehensive, making it easy for Claude to distinguish this skill from others. Minor improvement could be made by using slightly more natural language in the trigger terms section.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions: execute skill via ccw cli Claude, evaluate quality via Gemini, apply improvements via Agent, iterate until quality threshold or max iterations. The description names specific tools and a clear workflow.

3 / 3

Completeness

Clearly answers both 'what' (iterative skill tuning via execute-evaluate-improve feedback loop with specific tools) and 'when' (explicit triggers listed: 'skill iter tune', 'iterative skill tuning', 'tune skill').

3 / 3

Trigger Term Quality

Includes explicit trigger phrases ('skill iter tune', 'iterative skill tuning', 'tune skill') that users would naturally say. These are specific and varied enough to cover common phrasings.

3 / 3

Distinctiveness Conflict Risk

Highly distinctive niche combining skill tuning, iterative feedback loops, and specific tool chain (ccw cli, Gemini, Agent). The trigger terms are unique and unlikely to conflict with other skills.

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.

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

10

/

11

Passed

Repository
catlog22/Claude-Code-Workflow
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.