data-research

Structured data research: search sources, extract structured data, archive raw sources, maintain canonical tracker pages, deduplicate. Parameterized via YAML recipes for investor updates, donations, company updates, or any email-to-structured-data pipeline.

Quality

61%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/data-research/SKILL.md

Quality

Content

62%

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

The body is a well-sequenced, actionable 7-phase pipeline with explicit integrity safeguards and concrete tool usage. Weaknesses are mild verbosity in a few spots and a monolithic single-file structure that could offload recipe/extraction detail to referenced files.

Suggestions

Tighten the Phase 4 anecdote ('13/13 wrong ...') to a one-line rationale and remove the Built-In Recipes duplication of Phase 1's recipe list to improve conciseness.

Move extraction schema details and per-recipe field definitions into a referenced file (e.g., references/recipes.md) and link to it from Phase 1, so progressive disclosure can score 3.

Add a brief explicit 'validate tracker page parses before appending' checkpoint in Phase 7 to reinforce the existing integrity pattern across mutating steps.

Dimension	Reasoning	Score
Conciseness	Mostly lean with no padding of concepts Claude already knows, but the '13/13 wrong' anecdote and the Built-In Recipes section (which partially restates Phase 1) could be tightened; not every token earns its place.	2 / 3
Actionability	Gives concrete, executable guidance: specific tool calls (put_raw_data, file_upload, add_link), exact paths ('~/.gbrain/recipes/{name}.yaml'), a scaffolding command ('gbrain research init'), explicit dedup rules, and a copy-ready output table — matching the fully-actionable anchor.	3 / 3
Workflow Clarity	A clearly sequenced 7-phase pipeline with explicit validation (Phase 4 extraction-integrity rule: save raw, re-read from saved files) and a feedback loop (Phase 3 fail-improve logging), satisfying the anchor for clear sequence with validation and error-recovery loops.	3 / 3
Progressive Disclosure	Sections are well-organized and external references are minimal (skills/conventions/quality.md, recipe paths), but with no bundle files the recipe schemas and extraction details sit inline in a single ~110-line file rather than being split out, so structure is present but not deeply separated.	2 / 3
	Total	10 / 12 Passed

Description

60%

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 specific about capabilities and names a clear parameterized niche, but it omits an explicit 'Use when' trigger clause and relies on a separate triggers field for natural keyword coverage. Trigger guidance and distinctiveness are the main weaknesses.

Suggestions

Append an explicit 'Use when...' clause to the description (e.g., 'Use when the user asks to research, track, or extract structured data from email, web, or API sources') so completeness can reach 3.

Fold one or two high-signal natural trigger phrases ('track', 'build a tracker', 'extract from email') into the description body instead of relying solely on the triggers field.

Sharpen the niche framing to reduce overlap — lead with 'email-to-structured-data pipeline' rather than the generic 'Structured data research' opener.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — 'search sources, extract structured data, archive raw sources, maintain canonical tracker pages, deduplicate' — matching the multiple-specific-actions anchor rather than the single-action score-2 anchor.	3 / 3
Completeness	The 'what' is explicit (search/extract/archive/deduplicate), but there is no 'Use when...' clause; the 'when' is only implied via the example pipelines, which the rubric caps at 2.	2 / 3
Trigger Term Quality	Includes some natural terms ('research', 'investor updates', 'donations', 'company updates') but the core trigger verbs ('track', 'build a tracker', 'extract from email') live in the separate triggers field, so common variations are missing from the description itself.	2 / 3
Distinctiveness Conflict Risk	The 'email-to-structured-data pipeline' niche is somewhat specific, but generic verbs like 'research' and 'track' could overlap with other research or tracking skills, so it is not a clearly distinct, conflict-free niche.	2 / 3
	Total	9 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: garrytan/gbrain
Path: skills/data-research/SKILL.md
Commit: 3fafb69

Reviewed: about 21 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.