baseline

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

Quality

85%

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

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 highly actionable and well-structured, with a clear validated workflow and a real, well-signaled one-level reference bundle. Its main weakness is conciseness: pervasive restatement of the same comparator-first/stop-early guidance inflates the body and competes with context budget.

Suggestions

De-duplicate near-verbatim restatements (e.g., the repeated 'baseline should usually stop immediately' lines 153-154, and the recurring 'do not front-load a full codebase audit' / 'do not require a fresh memory pass' clauses) into a single canonical statement.

Consolidate the overlapping Constraints, Validation, and Hard acceptance gates sections, which restate the same comparability/contract requirements, to reclaim tokens without losing guidance.

Dimension	Reasoning	Score
Conciseness	The body avoids teaching concepts Claude already knows, but it is heavily padded with restated guidance — e.g., 'baseline should usually stop immediately' is repeated on adjacent lines (153-154) and 'do not front-load a full codebase audit' / 'do not require a fresh memory pass' recur near-verbatim — so it could be tightened well beyond 'mostly efficient'.	2 / 3
Actionability	Guidance is concrete and executable for an instruction skill: named artifact calls (artifact.confirm_baseline(...), artifact.waive_baseline(...), artifact.overwrite_baseline(...), artifact.arxiv(...)), a canonical contract path (<baseline_root>/json/metric_contract.json), explicit payload fields, and enumerated blocked-class and verification-verdict taxonomies.	3 / 3
Workflow Clarity	A numbered 5-step Control workflow with an explicit 'Verify before acceptance' checkpoint, a dedicated Verification section with outcome classification, and stop-rule feedback loops (failure class -> repair/blocked/waive) plus checklists (Validation, Hard acceptance gates, Exit criteria) match the anchor for clear sequencing with validation and error-recovery loops.	3 / 3
Progressive Disclosure	The body points to real one-level-deep references (route-selection.md, operational-guidance.md, artifact-flow-examples.md, boundary-cases.md, comparability-contract.md, artifact-payload-examples.md, baseline-plan-template.md, baseline-checklist-template.md — all present in ./references/) with 'read X' / 'use it when' navigation cues, and reference files do not chain into further references.	3 / 3
	Total	11 / 12 Passed

Description

85%

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 concrete and complete with an explicit 'Use when' trigger and a clear, distinctive niche. Its main weakness is trigger-term naturalness: the keywords are domain jargon with limited synonymous variation rather than the common phrasings a broad user would say.

Suggestions

Soften jargon and add natural-user phrasings (e.g., 'set up', 'fix', 'check', or 'reuse a baseline') alongside the formal verbs so the trigger matches what users actually say.

Add a few common variations of the core concept (e.g., 'reference run', 'comparison point') to broaden trigger coverage beyond the single term 'baseline'.

Dimension	Reasoning	Score
Specificity	Lists seven concrete actions — 'attach, import, reproduce, repair, verify, compare, or publish a baseline and its metrics' — matching the anchor for multiple specific concrete actions.	3 / 3
Completeness	An explicit 'Use when a quest needs to...' clause supplies the trigger while the verb list states what the skill does, satisfying both what and when with explicit triggers.	3 / 3
Trigger Term Quality	The verbs (reproduce, repair, verify) are relevant and domain-natural, but 'quest', 'baseline and its metrics' are specialized jargon and the description lacks common synonymous variations of a single concept, so it sits at 'some relevant keywords but missing common variations' rather than full coverage.	2 / 3
Distinctiveness Conflict Risk	The baseline-and-metrics niche within a staged quest workflow is distinct, and the specialized action set makes triggering for the wrong stage unlikely despite the mildly generic 'verify'/'compare' verbs.	3 / 3
	Total	11 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: ResearAI/DeepScientist
Path: src/skills/baseline/SKILL.md
Commit: b366244

Reviewed: 2 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.