ds-analysis-campaign

Use when a quest needs one or more follow-up runs such as ablations, robustness checks, error analysis, or failure analysis after a main experiment.

Quality

75%

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-analysis-campaign/SKILL.md

Quality

Content

77%

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

A highly actionable, clearly sequenced campaign-orchestration protocol with strong validation checkpoints and well-signaled one-level references, weakened mainly by length and repetition and by keeping detailed monitoring/palette/schema material inline rather than in separate reference files.

Suggestions

Extract the long bash_exec monitoring playbook (lines ~413-435) and the connector-chart palette guidance (lines ~44-59) into dedicated reference files, leaving SKILL.md as a leaner overview that links to them.

De-duplicate the PLAN.md/CHECKLIST.md creation requirement and the milestone-report content list, each of which is restated several times across the Interaction discipline, Quick workflow, Required plan and checklist, and Workflow sections.

Dimension	Reasoning	Score
Conciseness	It is substantive domain protocol with no padding of concepts Claude already knows, but at ~640 lines it restates the same rules repeatedly (PLAN.md/CHECKLIST.md requirements, bash_exec monitoring lists, milestone report contents) and could be tightened considerably; not a 1 because the content is genuinely non-obvious runtime protocol rather than fluff.	2 / 3
Actionability	Guidance is concrete and executable throughout — exact artifact calls ("artifact.create_analysis_campaign(...)", "artifact.record_analysis_slice(...)"), explicit bash_exec modes, required field lists, file paths, and a numbered monitoring cadence (60s/120s/300s/600s/1800s).	3 / 3
Workflow Clarity	A clearly sequenced Workflow (0 through 6) plus a Quick workflow, with explicit validation checkpoints ("do not mark a slice complete until the managed log and outputs both confirm completion", "verify the return path immediately") and feedback loops (smoke-test before real run, validate-fix-retry, stale-matrix re-open).	3 / 3
Progressive Disclosure	The five reference files are real, one level deep, and well-signaled with purpose (e.g. "Use references/campaign-plan-template.md as the canonical structure"), but the SKILL.md itself is a monolithic wall with long inline blocks — bash_exec monitoring details, connector-chart palette guidance, and per-slice field schemas — that would be better split into referenced files.	2 / 3
	Total	10 / 12 Passed

Description

72%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A specific, trigger-rich description with a clear explicit "Use when" clause and a well-bounded niche, but it leads entirely with the trigger and leaves the skill's own concrete actions implied rather than stated. Tightening it to name what the skill does would lift specificity and completeness.

Suggestions

Add an explicit capability statement naming the skill's actions (e.g. "Orchestrate and run coordinated follow-up analysis campaigns...") before the "Use when" clause so the "what" is stated, not implied.

Promote at least one follow-up type to a verb form (e.g. "run ablations", "check robustness") to convert run-type nouns into concrete actions for the specificity dimension.

Dimension	Reasoning	Score
Specificity	It names the domain and several concrete follow-up run types ("ablations, robustness checks, error analysis, or failure analysis"), but frames them as the quest's need ("a quest needs one or more follow-up runs such as...") rather than stating the skill's own actions, so it stops short of the verb-driven concrete-action list of a 3.	2 / 3
Completeness	The "when" is explicit ("Use when a quest needs one or more follow-up runs ... after a main experiment"), but the "what" is only implied through the listed run-types rather than stated as the skill's action, and the whole sentence is a trigger clause with no separate capability statement.	2 / 3
Trigger Term Quality	It surfaces natural terms a researcher would actually say — "ablations, robustness checks, error analysis, failure analysis, follow-up runs, main experiment" — giving good coverage of likely trigger phrasing.	3 / 3
Distinctiveness Conflict Risk	The niche is narrow and well-bounded — follow-up evidence runs after a main experiment — with distinct triggers (ablations/robustness/error/failure analysis) that are unlikely to fire for an unrelated skill.	3 / 3
	Total	10 / 12 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 (644 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-analysis-campaign/SKILL.md
Commit: 6da7f7c

Reviewed: about 7 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.