CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

mixed-check

Selected graph skill

Quality

56%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./crates/engine/tests/fixtures/extension_host/raw-dsh-mcp/source/skills/mixed-check/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

58%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.

The body is maximally lean (a strong 5 on conciseness and trivially well-structured for its size) but tells Claude nothing executable: what 'inspect' means, how to find the selected graph, or what result to produce is left entirely to guesswork. The single vague directive caps actionability and workflow clarity at the low end, making the skill unusable as written despite its token efficiency.

Suggestions

Define 'inspect' concretely: specify the exact command, API, or steps to run and the output to produce (e.g., a code snippet or command that dumps the selected graph's structure).

Explain how to locate 'the exact selected graph' (e.g., from editor state, a session variable, or a file path) so the directive is executable.

If inspection can fail or produce multiple candidates, add a short validation/checkpoint step so the workflow is unambiguous.

DimensionReasoningScore

Conciseness

The body 'Inspect the exact selected graph.' is five words with zero padding, no explanation of concepts Claude already knows, and no filler — every token earns its place. This matches anchor 5 ('Lean and efficient; assumes Claude's competence').

5 / 5

Actionability

'Inspect the exact selected graph.' is a single high-level directive with no method, tool, command, code, or expected output — matching anchor 2 ('Minimal concrete guidance; high-level hints but missing the specific steps to execute'). It is above anchor 1 because it instructs a specific action on a specific object rather than merely describing, and below anchor 3 because even pseudocode-level concreteness is absent: 'inspect' is left entirely undefined.

2 / 5

Workflow Clarity

The single action is stated but ambiguous — 'inspect' has no defined procedure, the expected output is unspecified, and how to identify 'the exact selected graph' is unaddressed (anchor 2: 'steps poorly defined'). The simple-skill exception would allow a 5 only if the single action were unambiguous, which it is not.

2 / 5

Progressive Disclosure

The skill is a single line under 50 lines with no bundle files (no references/, scripts/, or assets/ directories exist) and no content that belongs in separate files, so the simple-skill exception applies: the trivial structure is fully appropriate, with no monolithic text, buried references, or nested references.

5 / 5

Total

14

/

20

Passed

Description

32%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.

The description is a vague noun phrase that names a domain but no actions, no use conditions, and no synonyms, placing it in the same family as the rubric's bad examples like 'Processes data'. It would rarely trigger correctly and is indistinguishable from any other graph-related skill. Trigger-term coverage is the only dimension that rises slightly above poor, thanks to the single natural keyword 'graph'.

Suggestions

Rewrite the description to state concrete actions in third person, e.g. 'Inspects the currently selected graph, reporting its nodes, edges, and properties.'

Add an explicit trigger clause: 'Use when the user mentions a selected graph, chart inspection, or asks to examine the current graph.'

Include natural synonyms and related terms (chart, plot, visualization, network) to improve both trigger coverage and distinctiveness from other graph-related skills.

DimensionReasoningScore

Specificity

The description 'Selected graph skill' names a domain ('graph') but contains no action verbs whatsoever — a bare noun phrase. It sits between anchor 1 ('no concrete actions; pure abstract language', e.g. 'Helps with documents') and anchor 2 ('names the domain but actions are minimal or generic', e.g. 'Processes PDF files'): a domain is named, but the action coverage is zero rather than minimal.

2 / 5

Completeness

There is only a vague 'what' ('Selected graph skill' implies some graph-related capability but names no actions) and no 'when' clause at all — an exact match for anchor 2 ('Has a vague what and no when'). The missing 'Use when...' guidance caps it well below 4 per the judging guidelines.

2 / 5

Trigger Term Quality

'graph' is a single relevant, natural keyword a user might say, matching anchor 3's pattern ('Works with PDF files'), but every common variation and synonym is missing ('chart', 'plot', 'visualization', 'network'). It is not anchor 2 because 'graph' is domain-relevant rather than generic like 'files', and not anchor 4 because coverage is one keyword with no synonyms or extensions.

3 / 5

Distinctiveness Conflict Risk

'graph' alone is ambiguous across chart-visualization, graph-theory, and knowledge-graph skills, creating high overlap risk with many similar skills (anchor 2). It does not reach anchor 3 because nothing in the description narrows the scope the way 'selected' plus a concrete action could.

2 / 5

Total

9

/

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

Hmbown/Codewhale

Path

crates/engine/tests/fixtures/extension_host/raw-dsh-mcp/source/skills/mixed-check/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.