CtrlK
BlogDocsLog inGet started
Tessl Logo

ark-research

Research technical solutions by searching the web, examining GitHub repos, and gathering evidence. Use when exploring implementation options or evaluating technologies.

79

1.65x
Quality

79%

Does it follow best practices?

Impact

68%

1.65x

Average score across 3 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/research/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 well-structured, actionable research workflow with executable commands, a concrete output template, an evidence threshold, and fallback handling for blocked content. Weaknesses are modest: minor redundant framing, some high-level steps lacking concrete mechanics, and inline content (output template, worked example) that could be split into reference files for progressive disclosure.

Suggestions

Trim redundant framing — the opening line 'Research technical solutions and gather evidence before implementation' duplicates the frontmatter description, and 'Web Search First' can drop the 'Always start with' preamble since the heading already conveys ordering.

Make the search and comparison steps concrete: give example query patterns for the initial web search and specify what a comparison note in ./scratch/research/ should contain (criteria, tradeoffs, verdict per option).

Move the Output Format template and Example Usage into a references/ file (e.g., references/output-format.md) referenced one level deep, keeping SKILL.md as a lean overview.

DimensionReasoningScore

Conciseness

The body is efficient — no explanations of concepts Claude already knows, and lists ('Official documentation', 'README documentation, Code examples') carry real guidance. Minor trimmable padding: the intro line 'Research technical solutions and gather evidence before implementation' restates the frontmatter description, and 'Always start with web search' duplicates the section title's intent. Fits anchor 4 (efficient, minor over-explanation) rather than 5 (every token earns its place); well above 3.

4 / 5

Actionability

Provides executable commands (`git clone https://github.com/owner/repo.git`, `mkdir -p ./scratch/research`), a copy-paste-ready output template, a concrete evidence threshold ('Minimum 2-3 datapoints required'), and a worked example. Not 5 because key steps remain high-level — no concrete command or approach for the web-search step itself or for summarizing cloned repos — leaving minor gaps per anchor 4.

4 / 5

Workflow Clarity

A clear five-step numbered sequence with error recovery (section 3, 'Handle Blocked Content', gives a ready-made fallback prompt) and an evidence-count checkpoint with an escalation path ('If insufficient evidence, ask for guidance'). Not 5 because the evidence check is a threshold rather than an explicit validate-fix-retry loop, and steps like 'Compare approaches' lack a stated checkpoint. Not a destructive/batch operation, so no cap applies.

4 / 5

Progressive Disclosure

No bundle files exist, and the ~100-line body is organized into well-signaled sections (Research Process, Output Format, Example Usage). Falls short of 5 because the skill exceeds the under-50-line simple-skill exception and inlines content that could live in a reference file — notably the ~20-line Output Format template and the Example Usage walkthrough. Comfortably above 3: structure is clear and nothing is buried.

4 / 5

Total

16

/

20

Passed

Description

83%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 description: concrete third-person actions plus an explicit 'Use when...' clause with natural trigger phrases. Main weakness is missing common synonyms (e.g., 'research', 'compare libraries') that users would likely say, and slight overlap risk with general research/web-search skills.

DimensionReasoningScore

Specificity

Quotes several concrete actions — 'searching the web, examining GitHub repos, and gathering evidence' — but 'gathering evidence' is generic, leaving minor gaps in coverage. Fits anchor 4 ('lists several specific actions; minor gaps') better than 5, whose example enumerates a comprehensive list of concrete operations, and better than 3, which covers only 1-2 actions.

4 / 5

Completeness

Explicitly answers what ('Research technical solutions by searching the web, examining GitHub repos, and gathering evidence') and when ('Use when exploring implementation options or evaluating technologies') with concrete trigger phrases. Matches the 5 anchor structure; not 4 because the 'when' clause is explicit and specific rather than generic like 'Use when working with PDF files'.

5 / 5

Trigger Term Quality

Natural terms like 'searching the web', 'GitHub repos', 'implementation options', 'evaluating technologies' are present and would be said by users, but common synonyms such as 'research', 'compare libraries/frameworks', or 'technical due diligence' are missing. Not 5 (no synonym/extension-level coverage like 'PDFs, .pdf'), clearly above 3 (keywords are relevant and varied, not just domain-naming).

4 / 5

Distinctiveness Conflict Risk

The GitHub/web-research framing carves a fairly distinct niche, but 'Research technical solutions' and 'evaluating technologies' overlap with generic research or web-search skills. Minor overlap risk with closely related skills fits anchor 4; not 5 because triggers are less uniquely scoped than a format-specific skill, and not 3 because the description is far from broadly generic.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 3 missing

Warning

Total

15

/

16

Passed

Repository
mckinsey/agents-at-scale-ark
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.