ds-baseline

Use when a quest needs to attach, import, reproduce, repair, verify, compare, or publish a baseline and its metrics.

Quality

88%

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

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 thorough, highly actionable baseline-stage skill with strong workflow sequencing and concrete tool calls. Its weak spots are token-efficiency (redundant sections that could be tightened) and progressive disclosure (an over-long overview plus two undiscoverable reference files).

Suggestions

Deduplicate the gate/exit condition lists (Stage gate, Exit criteria, Required durable outputs) into a single canonical statement and reference it, and merge the overlapping bash_exec rules in Interaction discipline and Tool discipline.

Move the inline Durable path contract and Metric-contract rules into a reference file so SKILL.md reads as an overview, then link it alongside the existing references.

Link references/codebase-audit-checklist.md and references/memory-playbook.md from the relevant sections (escalation-to-fuller-audit and Memory rules) so all bundle files are discoverable.

Dimension	Reasoning	Score
Conciseness	The body assumes Claude's intelligence (no basic-concept padding) but carries real redundancy — the two "Default to a fast path..." sentences near-duplicate (lines 73-74), gate/exit conditions recur across Stage gate, Exit criteria, and Required durable outputs, and the bash_exec discipline is stated in both Interaction and Tool discipline sections.	2 / 3
Actionability	It gives concrete, copy-paste-ready guidance throughout: uv patterns ("uv sync", "uv venv --python 3.11", "uv run python scripts/smoke_test.py"), explicit tool calls ("bash_exec(mode='detach', ...)", "artifact.confirm_baseline(...)"), and precise path contracts ("<quest_root>/baselines/local/<baseline_id>/").	3 / 3
Workflow Clarity	A clear four-phase sequence (analysis -> setup -> execution -> verification) is paired with explicit validation checkpoints (smoke test before real run, mandatory verification before acceptance) and bounded retry feedback loops plus a required CHECKLIST.md.	3 / 3
Progressive Disclosure	Linked references are well-signaled and one level deep, but the SKILL.md is heavier than an overview (inline durable-path and metric-contract material that reads like reference content) and 2 of 8 bundle files (codebase-audit-checklist.md, memory-playbook.md) are not linked anywhere, leaving a navigation gap.	2 / 3
	Total	10 / 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.

A concise, well-formed description that names concrete actions, carries an explicit Use-when trigger, and stakes out a distinct baseline niche. It scores at the top of the scale across all four dimensions without over-claiming or padding.

Dimension	Reasoning	Score
Specificity	The description lists seven concrete actions ("attach, import, reproduce, repair, verify, compare, or publish") applied to a concrete object ("a baseline and its metrics"), matching the multiple-specific-actions anchor.	3 / 3
Completeness	It explicitly answers both what (the route verbs on a baseline and its metrics) and when via the leading "Use when a quest needs to..." clause, satisfying the explicit-trigger requirement.	3 / 3
Trigger Term Quality	It surfaces natural ML-research trigger terms ("baseline", "metrics", "reproduce", "repair", "compare", "publish") alongside an explicit "Use when" trigger, giving good coverage of words a user would actually say.	3 / 3
Distinctiveness Conflict Risk	The baseline-and-metrics niche plus distinctive route verbs (attach/import/reproduce/repair) carve a clear scope unlikely to collide with sibling stage skills.	3 / 3
	Total	12 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (609 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	13 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/ds-baseline/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.