research-curator

Orchestrate research entry lifecycle in ./research/ — create, batch-import, refresh stale entries, and validate structure. Use when asked to add a tool, research a URL, document a library, refresh research, validate entries, or given any tool or library URL. Supports --batch (parallel multi-URL), --rerun (refresh one or all entries), and --validate (structural check with auto-fix of error-severity issues).

Quality

85%

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

Quality

Content

81%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.

A highly actionable, well-structured multi-mode orchestrator with strong validation feedback loops and verified one-level-deep references. Its main weakness is conciseness: the validation gate and Mermaid diagrams are duplicated across modes where a shared reference would do.

Suggestions

Extract the repeated validation-gate block (fix_research_formatting.py + validate_research.py main --json, plus the pass/fail branching) into a single shared section or reference file and point each mode to it instead of restating it four times.

For each mode, keep either the Mermaid diagram or the numbered step list as the authoritative procedure — not both — to remove the diagram/prose duplication and tighten the token budget.

Consider moving the cross-cutting Agent Result Relay Rules and Output Format blocks into ./references/ files referenced once, since they apply identically to all modes.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining concepts Claude already knows, but the validation-gate block (fix script + validator invocation) is repeated verbatim across Default, Batch, Rerun-single, and Rerun-all modes, and each Mermaid diagram restates the adjacent numbered steps — padding that could be deduplicated.	3 / 5
Actionability	Provides copy-paste-ready commands (uv run .../validate_research.py main --json {file}, git push -u origin HEAD, prek run --files ...), exact Agent-tool prompt strings, and per-mode commit message templates covering the common cases.	5 / 5
Workflow Clarity	Each mode has an explicit validation gate with feedback loops (errors → mark "created/refreshed with issues" and skip analysis; pass → proceed), decision-point Mermaid diagrams, and a pre-relay quality checklist — the feedback-loop pattern the rubric rewards, present even for the batch/destructive operations.	5 / 5
Progressive Disclosure	Well-signaled one-level-deep references (entry-template, validation-rules, batch-mode, duplicate-detection — all verified present) with clear mode-tagged sections, but cross-cutting relay rules and output-format blocks are inlined rather than split, and diagram/prose duplication leaves minor organization gaps.	4 / 5
	Total	17 / 20 Passed

Description

88%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.

A strong, third-person description that crisply states both capability and explicit activation triggers with concrete flag-level detail. Its only soft spot is trigger-term breadth (no file extensions) and minor overlap risk with generic research skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "create, batch-import, refresh stale entries, and validate structure" — plus per-flag capabilities (--batch parallel multi-URL, --rerun, --validate with auto-fix), giving comprehensive coverage with no real gaps.	5 / 5
Completeness	Explicitly answers "what" (orchestrate the research entry lifecycle: create, batch-import, refresh, validate) and "when" ("Use when asked to add a tool, research a URL, document a library, refresh research, validate entries, or given any tool or library URL") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural phrases users would say ("add a tool", "research a URL", "document a library", "refresh research", "validate entries") with synonyms across tool/library/URL, but lacks file extensions or further synonyms that would push it to 5.	4 / 5
Distinctiveness Conflict Risk	The ./research/ entry-lifecycle framing is a clear, mostly distinct niche, but "research a URL" and "given any tool or library URL" carry minor overlap risk with general web-research skills rather than the minimal risk of a 5.	4 / 5
	Total	18 / 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 (527 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: Jamie-BitFlight/claude_skills
Path: .claude/skills/research-curator/SKILL.md
Commit: 4e1e73b

Reviewed: ~13 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.