CtrlK
BlogDocsLog inGet started
Tessl Logo

judge

Launch a meta-judge then a judge sub-agent to evaluate results produced in the current conversation

53

Quality

60%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./plugins/sadd/skills/judge/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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 content is a well-sequenced, actionable evaluation pipeline with concrete prompt templates and validation steps. Its main weakness is redundancy between the workflow, the 13-item guidelines, and the notes, which inflates token cost without adding proportional value.

Suggestions

Fold the 'Important Guidelines' and 'Notes' sections into the relevant workflow phases to eliminate restatement and reduce length.

Make Phase 4 validation items more executable (e.g., specific checks for score-range validity and weighted-total arithmetic rather than vague 'verify' instructions).

Consider moving the large prompt templates into a references/ file referenced from the body to improve progressive disclosure for a 200-line skill.

DimensionReasoningScore

Conciseness

The body is mostly efficient procedural guidance, but the 13-item 'Important Guidelines' and 'Notes' sections restate concepts already covered in the workflow phases, so it could be tightened.

3 / 5

Actionability

Provides concrete, mostly executable guidance — full meta-judge and judge prompt templates plus Task-tool dispatch blocks with model and subagent_type — though prompts rely on placeholders and reference external agent instructions.

4 / 5

Workflow Clarity

The four-phase pipeline is clearly sequenced and Phase 4 includes an explicit validation checklist with a re-evaluation feedback loop, but the validation items are high-level checks rather than executable verification commands.

4 / 5

Progressive Disclosure

Well-organized into clear sections (workflow, phases, scoring table, guidelines, notes) with no external bundle files needed; a 200-line single-file skill with good structure but no references to split out.

4 / 5

Total

15

/

20

Passed

Description

53%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 states what the skill does reasonably specifically but lacks any explicit 'when to use it' trigger guidance, capping completeness. Trigger terms are present but partly technical and missing common synonyms.

Suggestions

Add an explicit 'Use when...' clause naming natural trigger phrases (e.g., 'Use when the user asks to review, assess, or evaluate work produced in this conversation').

Replace or supplement jargon like 'meta-judge' with user-facing synonyms such as 'review' or 'assess' so the description matches what users actually say.

Briefly mention the output (a report-only evaluation with structured scores) so the 'what' is more complete.

DimensionReasoningScore

Specificity

Names the evaluation domain and two concrete actions ('Launch a meta-judge then a judge sub-agent to evaluate results'), but coverage is not comprehensive — it omits what the judge actually produces or how results are surfaced.

3 / 5

Completeness

Has a clear 'what' but no 'Use when...' clause or equivalent trigger guidance, which per the judging guidelines caps completeness at 3.

3 / 5

Trigger Term Quality

Contains some relevant keywords ('evaluate results', 'judge') but leans on jargon ('meta-judge', 'judge sub-agent') and misses common natural synonyms users would say like 'review', 'assess', or 'critique'.

3 / 5

Distinctiveness Conflict Risk

The meta-judge -> judge two-phase pattern is a fairly distinct niche with minimal overlap risk against generic review skills, though 'evaluate results' alone is somewhat broad.

4 / 5

Total

13

/

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.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
NeoLabHQ/context-engineering-kit
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.