inno-experiment-dev

Creates implementation plan, writes project code with judge feedback loop, and submits final experiment run. Use after code-survey in both Idea and Plan branches.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/inno-experiment-dev/SKILL.md

Quality

Content

85%

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

The body is well-organized, concise, and workflow-driven with strong progressive disclosure and validation checkpoints; the main gap is that actionable guidance is expressed as orchestration prose rather than executable code snippets.

Suggestions

Add one or two concrete code snippets (e.g. the exact messages-list construction or a sample build_plan_query call) to lift actionability from orchestration prose to copy-ready.

Consider including a short example of the judge iteration loop's message-appending and early-exit logic so the feedback loop is executable rather than described.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — no conceptual padding about ML/training, every section (inputs, cache artifacts, tool mappings, checklist) earns its place with high signal density.	3 / 3
Actionability	Guidance is concrete — exact prompt-builder signatures, message-list formats, cache paths, and line-number references — but it is orchestration prose rather than copy-paste executable code, so it stops short of fully copy-ready examples.	2 / 3
Workflow Clarity	A clear three-phase numbered sequence with an iteration loop, explicit early-exit on 'fully_correct', a verification checkpoint (Phase 1 step 4, Phase 3 step 11), and a closing checklist with feedback-driven error recovery.	3 / 3
Progressive Disclosure	SKILL.md serves as an overview pointing to three real one-level-deep references (coding_plan_agent.md, ml_agent_instructions.md, judge_agent_instructions.md), each clearly signaled inline and confirmed to exist in the bundle.	3 / 3
	Total	11 / 12 Passed

Description

70%

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 is specific, complete, and clearly distinguishable, but its trigger clause relies on internal pipeline stage names rather than natural user language, weakening trigger-term quality.

Suggestions

Add natural-language trigger terms a user might actually say (e.g. 'implementation planning', 'iterate with judge feedback', 'submit experiment run') alongside the pipeline-stage references.

Keep the explicit 'Use when/after...' clause but broaden it so it reads as a user-facing trigger rather than only an internal pipeline dependency.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Creates implementation plan, writes project code with judge feedback loop, and submits final experiment run' — naming the plan, code-writing, feedback loop, and submission steps.	3 / 3
Completeness	Explicitly answers both what it does (plan, code, feedback loop, submit) and when to use it ('Use after code-survey in both Idea and Plan branches'), satisfying the explicit-trigger requirement.	3 / 3
Trigger Term Quality	Triggers reference internal pipeline jargon ('Use after code-survey in both Idea and Plan branches') rather than keywords a user would naturally say; no natural-language trigger terms are present.	1 / 3
Distinctiveness Conflict Risk	Targets a clear, narrow niche (the inno experiment dev pipeline) with a specific positional trigger ('after code-survey'), making it unlikely to fire for unrelated skills.	3 / 3
	Total	10 / 12 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: OpenLAIR/dr-claw
Path: skills/inno-experiment-dev/SKILL.md
Commit: 6da7f7c

Reviewed: about 8 hours ago

Table of Contents

Discovery Implementation Validation

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.