CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

research

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

An exemplary lean, instruction-only skill: it assumes Claude's competence, gives a fully concrete job spec for the background agent including source-quality criteria, output format, and placement convention, and handles the no-convention fallback explicitly. The only minor gap is the absence of a sample findings-file layout or citation format, which would make the output expectations copy-paste concrete.

DimensionReasoningScore

Conciseness

Every phrase changes behavior — "Start it with a fresh context window", "a single Markdown file", "citing each claim's source", "match the existing convention" — with zero padding and no explanation of concepts Claude already knows. This matches the lean, every-token-earns-its-place anchor.

5 / 5

Actionability

For an instruction-only skill the guidance is concrete and directive: spin up a background agent, investigate primary sources enumerated as "official docs, source code, specs, first-party APIs", write one Markdown file with per-claim citations, and place it per repo convention with an explicit fallback. It falls just short of anchor 5 because no example of the findings-file structure or citation format covers the common case.

4 / 5

Workflow Clarity

The sequence is clear and unambiguous — spin up the agent, then a numbered job spec — and includes an explicit decision branch ("if there is none, put it somewhere sensible and say where"). This is a single-purpose, non-destructive task, so the simple-skill exception applies and no validation checkpoint is required.

5 / 5

Progressive Disclosure

The skill is under 50 lines, needs no external references, and no bundle files exist — the body is well organized into a setup line, a job header, and a numbered list. Per the simple-skill guideline, this earns the top score without external file references.

5 / 5

Total

19

/

20

Passed

Description

78%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 well-crafted description with an explicit 'Use when...' clause covering multiple concrete trigger phrases and a clear two-part statement of what the skill does. Its only weakness is moderate action coverage — it names two concrete actions rather than several, and could include a few more natural phrasing variations users might say.

DimensionReasoningScore

Specificity

Lists two concrete actions — "Investigate a question against high-trust primary sources" and "capture the findings as a Markdown file in the repo" (with output format and location specified). This matches the anchor for 1-2 concrete actions; it falls short of anchor 4, which requires several specific actions, and is clearly above anchor 2's generic actions.

3 / 5

Completeness

Explicitly answers both: what ("Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo") and when ("Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent") with concrete trigger phrases. This matches the anchor-5 example pattern; anchor 4's weaker 'when' clause does not apply.

5 / 5

Trigger Term Quality

Natural phrases users would say are present: "a topic researched", "docs or API facts gathered", "reading legwork delegated". Good coverage, but common variations like "look up", "find out about", or "check the documentation" are missing, so it does not reach comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

The niche — delegated primary-source research captured as a repo Markdown note — is mostly distinct with specific triggers. Minor overlap risk remains with generic web-search or documentation-lookup skills on broad terms like "research" and "docs", keeping it below anchor 5 but well above anchor 3.

4 / 5

Total

16

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

mattpocock/skills

Path

skills/engineering/research/SKILL.md

Reviewed

Table of Contents

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.