CtrlK
BlogDocsLog inGet started
Tessl Logo

derive-review-rubrics

Derive custom review-plugin judges, rubrics, and scoring criteria from evidence — existing skills, PR review feedback, and accumulated agent logs — by finding where agents needed correction or skills failed to activate, then translating those patterns into scoring dimensions and anchors. Hands the rubric design to create-review-plugin for scaffolding. Use when the user wants a custom skill reviewer or scoring rubric grounded in how their agents actually behave — to grade skills consistently, or build review criteria from past PR feedback and agent logs.

89

1.31x
Quality

80%

Does it follow best practices?

Impact

95%

1.31x

Average score across 6 eval scenarios

SecuritybySnyk

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./review-plugin-creator/skills/derive-review-rubrics/SKILL.md

The canonical home for this skill is tessl/review-plugin-creator

SKILL.md
Quality
Evals
Security

Quality

Content

67%Scale 1-5

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

This is a well-structured orchestration skill that clearly defines a three-step evidence-gathering and rubric-design workflow. Its main strengths are clear workflow sequencing, good boundary definition (explicitly stating what it does and does not do), and appropriate delegation to other skills. Its weaknesses are moderate verbosity in edge-case explanations and the lack of a concrete template or schema for the design file output that gets handed to create-review-plugin.

Suggestions

Add a concrete example or template of the design file output that gets written in Step 3 — showing the expected structure with judges, dimensions, weights, and grounded anchors would make the handoff much more actionable.

Tighten the agent-logs section by consolidating the empty-entries vs no-history distinction into a brief decision table rather than prose paragraphs.

DimensionReasoningScore

Conciseness

The skill is moderately efficient but includes some unnecessary elaboration. For example, the detailed explanation of how Claude Code expands slash commands, the distinction between empty entries vs empty content, and the explanation of what 'entries present but with empty content' means could be tightened. However, much of the content is domain-specific knowledge Claude wouldn't inherently know.

3 / 5

Actionability

The skill provides concrete commands (e.g., `tessl agent-logs view --json --since <ISO timestamp>`, `/find-optimizations`, `/create-review-plugin`) and specific patterns to look for in transcripts (tool_use blocks named `Skill`, slash commands starting with `/<skill>`). However, it lacks executable code examples and the output format of the design file is not specified with a concrete template or schema.

4 / 5

Workflow Clarity

The three-step workflow (Gather → Map → Hand off) is clearly sequenced with well-defined substeps. The skill includes good conditional logic (e.g., handling empty logs, distinguishing empty entries from no history). However, there are no explicit validation checkpoints between steps — for instance, no verification that evidence gathering produced sufficient signal before proceeding to rubric mapping, though the operation is not destructive.

4 / 5

Progressive Disclosure

The skill references `references/evidence-to-rubric.md` for detailed rubric construction rules and `.tessl/memory/improvement-types.md` for configurable types, keeping the main body at an appropriate overview level. However, since no bundle files were provided, we cannot verify these references exist or are well-structured. The references are one-level deep and clearly signaled, which is good practice.

4 / 5

Total

15

/

20

Passed

Description

92%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.

This is a strong, well-crafted description that clearly articulates a specific and distinctive capability. It provides comprehensive detail on what the skill does (derive rubrics from evidence patterns), how it works (analyzing agent corrections and skill activation failures), and when to use it (explicit 'Use when' clause with natural trigger phrases). The description is appropriately detailed without being verbose.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions: deriving judges/rubrics/scoring criteria from evidence, finding where agents needed correction or skills failed to activate, translating patterns into scoring dimensions and anchors, and handing off to create-review-plugin. This is comprehensive and specific.

5 / 5

Completeness

Clearly answers both 'what' (derive custom review-plugin judges, rubrics, and scoring criteria from evidence by finding correction patterns and translating them into dimensions/anchors) and 'when' (explicit 'Use when' clause with concrete triggers: custom skill reviewer, scoring rubric, grading skills, building review criteria from PR feedback and agent logs).

5 / 5

Trigger Term Quality

Includes strong natural terms like 'custom reviewer', 'scoring rubric', 'PR review feedback', 'agent logs', 'grade skills', 'review criteria'. Could benefit from a few more synonyms (e.g., 'evaluation criteria', 'quality rubric', 'linting') but coverage is good.

4 / 5

Distinctiveness Conflict Risk

Highly distinctive niche — deriving review rubrics from agent behavior evidence. The explicit mention of 'review-plugin judges', 'scoring dimensions and anchors', 'agent logs', and handoff to 'create-review-plugin' makes this clearly distinguishable from general code review, skill creation, or other skills.

5 / 5

Total

19

/

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.

Validation11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository
tesslio/product-plugins
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.