Name: evaluation
Rating: 51.2 (1 reviews)
Author: administrakt0r

evaluation

Build evaluation frameworks for agent systems. Use when testing agent performance systematically, validating context engineering choices, or measuring improvements over time.

Quality

56%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills/skills/evaluation/SKILL.md

The canonical home for this skill is evaluation in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

42%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 conceptually thorough but verbose, over-explaining fundamentals Claude already knows and relying on pseudocode rather than executable guidance. Structure exists but conceptual bulk that belongs in reference files is inlined, and the cited reference file is missing.

Suggestions

Trim conceptual explanations of agent non-determinism, composite quality dimensions, and the BrowseComp 95% finding to lean bullet points, since Claude already knows these fundamentals.

Replace the pseudocode evaluate_agent_response example with executable code that defines or imports load_rubric, assess_dimension, and weighted_average, and provide a runnable rubric-assessment snippet.

Either create the referenced Metrics Reference file under references/ and link it with a clear path, or remove the dangling "Metrics Reference" citation so navigation is not broken.

Dimension	Reasoning	Score
Conciseness	The body extensively explains concepts Claude already knows (e.g., "Agents make dynamic decisions, are non-deterministic between runs", the 95% BrowseComp finding, and long conceptual prose on non-determinism and composite quality), constituting several padded sections of unnecessary explanation.	2 / 5
Actionability	Two code examples are present but incomplete: evaluate_agent_response calls undefined load_rubric/assess_dimension/weighted_average functions (pseudocode), and the 8-step "Building Evaluation Frameworks" list is abstract guidance, fitting the anchor of some concrete guidance with missing executable details.	3 / 5
Workflow Clarity	The "Building Evaluation Frameworks" section lists an 8-step sequence, but steps are abstract with no validation checkpoints or feedback loops, matching the anchor of a present sequence with missing or implicit checkpoints.	3 / 5
Progressive Disclosure	Section headers give some structure, but large conceptual blocks that belong in separate reference files are inlined, and the "References" section cites a "Metrics Reference" that has no corresponding bundle file, so references are present but not clearly backed or signaled.	3 / 5
	Total	11 / 20 Passed

Description

70%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 cleanly answers both what the skill does and when to use it with natural trigger phrases and a distinct niche. Its main weakness is that the "what" is a single high-level action rather than a comprehensive list of concrete capabilities.

Dimension	Reasoning	Score
Specificity	"Build evaluation frameworks for agent systems" names the domain plus one concrete action (build), while the rest of the description lists use-cases rather than additional concrete capabilities, matching the anchor that names domain and 1-2 actions but is not comprehensive.	3 / 5
Completeness	It provides a clear "what" ("Build evaluation frameworks for agent systems") and an explicit "Use when..." clause with three concrete triggers, but the "what" is a single high-level action so it falls just short of the fully comprehensive anchor 5.	4 / 5
Trigger Term Quality	Natural phrases like "testing agent performance systematically", "validating context engineering choices", and "measuring improvements over time" are terms a user would plausibly say, giving good keyword coverage with only minor synonyms missing.	4 / 5
Distinctiveness Conflict Risk	"Build evaluation frameworks for agent systems" carves a fairly distinct niche with specific triggers, with only minor overlap risk against general testing or metrics skills.	4 / 5
	Total	15 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills/skills/evaluation/SKILL.md
Commit: 57c135b

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.