ds-experiment

Use when a quest is ready for a concrete implementation pass or a main experiment run tied to a selected idea and an accepted baseline.

Quality

60%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/ds-experiment/SKILL.md

Quality

Content

66%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.

The body is highly actionable and exceptionally well-sequenced with strong validation and feedback loops, but it is verbose: many rules are restated across sections and sizable detail blocks are inlined rather than moved to reference files. Tightening redundancy and offloading the chart and protocol detail would raise conciseness and progressive disclosure.

Suggestions

Dedupe the repeated bash_exec mandate, wait-cadence list, seven experiment fields, and evaluation_summary fields so each rule appears once.

Move the connector-chart palette and requirements and the full bash_exec long-running protocol into reference files, keeping the body to a short summary with a link.

Add one fully worked artifact.record_main_experiment(...) example call so the actionability reaches copy-paste readiness.

Dimension	Reasoning	Score
Conciseness	The ~800-line body restates the same rules in multiple places (the bash_exec mandate, the 60s/120s/300s wait cadence, the seven experiment fields, the evaluation_summary fields) and inlines detail such as chart palettes that should live in reference files, making it noticeably verbose with several padded sections.	2 / 5
Actionability	Provides concrete executable guidance via specific tool calls with parameters (artifact.record_main_experiment, bash_exec(mode='detach', ...), memory.list_recent), exact file paths, and explicit field schemas, with only minor gaps such as no fully worked example call.	4 / 5
Workflow Clarity	A clearly numbered Workflow (contract, preflight, implement, execute, validate, record, decide) is backed by explicit validation checkpoints (preflight, pilot-before-scale, acceptance gate), feedback loops (diagnostic mode, last-known-good, diagnosis loop), and referenced checklist templates.	5 / 5
Progressive Disclosure	References three real, one-level-deep reference files with clear signals (evidence-ladder, plan template, checklist template), but a fair amount of detail that could be split out (chart palettes, the full bash_exec long-running protocol, memory rules) is inlined in the main body.	4 / 5
	Total	15 / 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 gives a clear and specific trigger condition but only implies what the skill actually does, leaning on system jargon ('quest') instead of natural user phrasings. It is reasonably distinct within its workflow family.

Suggestions

State the concrete actions the skill performs (e.g., 'plan, run, validate, and record a main experiment, then route the next action') instead of only naming the stage.

Add natural user-facing trigger terms a person would actually say rather than system-internal terms like 'quest' and 'implementation pass'.

Include an explicit 'Use when...' clause that already exists, but pair it with a one-clause 'what' so both what and when are explicitly answered.

Dimension	Reasoning	Score
Specificity	Names the domain and two high-level activities ('concrete implementation pass', 'a main experiment run') but does not enumerate concrete actions like recording runs, validating outputs, or routing the next move.	3 / 5
Completeness	The 'when' is explicit ('Use when a quest is ready for...') but the 'what' is only implied by 'experiment run' with no stated actions, so the missing explicit 'what' keeps it from scoring higher.	3 / 5
Trigger Term Quality	Contains relevant domain terms ('experiment run', 'baseline', 'selected idea') but they are system-internal jargon ('quest') rather than natural phrases a user would say, and common variations are missing.	3 / 5
Distinctiveness Conflict Risk	The gating conditions (selected idea plus accepted baseline) carve a mostly distinct niche with only minor overlap risk against adjacent optimize or analysis stages.	4 / 5
	Total	13 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (811 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/ds-experiment/SKILL.md
Commit: 4b35756

Reviewed: 5 days 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.