CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

aip-knowledge

This skill should be used when the user asks about "AIP rules", "API Improvement Proposals", "Google API guidelines", "AIP-158", "AIP-193", or any specific AIP number. Also use when user asks "how should I implement pagination", "what's the right error format", "how do I design a REST API following Google's standards", or needs guidance on errors, pagination, filtering, field masks, long-running operations, or batch operations in REST/OpenAPI APIs.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/baume/skills/aip-knowledge/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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 lean, highly actionable quick-reference body whose inline tables and JSON examples cover the most common AIP patterns well. Its critical defect is that the entire progressive-disclosure layer is fictional: the body routes users to eight reference files and a lookup agent, none of which exist in the bundle, so detailed-guidance requests dead-end at step 2 of the skill's own workflow (partially rescued by the google.aip.dev URL).

Suggestions

Ship the referenced bundle files (errors.md, pagination.md, filtering.md, lro.md, field-masks.md, batch.md, rest-mapping.md, linter-rules.md) or remove the Reference Files table — the skill's core routing instruction currently fails because none of them exist.

Replace or explain the `baume-lookup` agent: either include it in the bundle/agent definitions with invocation instructions, or drop it in favor of the already-present direct fetch from https://google.aip.dev/{number}.

Until reference files exist, promote the "Fetch AIPs On Demand" section to the primary detailed-guidance path so the escalation workflow never routes to a missing file.

DimensionReasoningScore

Conciseness

The body is almost entirely dense reference material — routing table, method-to-HTTP mapping, naming do/don't pairs with rule IDs, and minimal copy-paste JSON snippets. It explains nothing Claude already knows and every line (e.g., the footnote "*Use Idempotency-Key header for safe retries") earns its place, matching the anchor 5 'lean and efficient' profile.

5 / 5

Actionability

Concrete, usable guidance throughout: exact query parameters ("?page_size=20&page_token=xxx"), a complete copy-paste error JSON schema, exact HTTP method/path/idempotency mappings, and naming examples tied to specific linter rules. It falls short of anchor 5 ("fully executable") because the primary instruction "Load the relevant reference file from the table" cannot be executed — none of the eight referenced files exist in the bundle.

4 / 5

Workflow Clarity

The "How to Use This Skill" section gives an unambiguous escalation path (quick patterns → reference file by topic → linter rules → deeper explanation via agent/google.aip.dev), which suits a reference skill with no multi-step process. It sits at anchor 4 rather than 5 because the `baume-lookup` agent is named with no instruction for how to invoke it, leaving a minor execution gap.

4 / 5

Progressive Disclosure

The in-body structure is exemplary — a topic/file/when-to-use table with one-level-deep, clearly signaled references — but scored against the actual bundle, every referenced path (errors.md, pagination.md, filtering.md, lro.md, field-masks.md, batch.md, rest-mapping.md, linter-rules.md, and the baume-lookup agent) is absent, so navigation dead-ends past the Quick Reference. The well-organized inline Quick Reference and the working google.aip.dev fetch URL keep this at anchor 3 rather than 2, but "easy navigation" (anchor 4/5) is not achieved when all nine referenced targets are missing.

3 / 5

Total

16

/

20

Passed

Description

68%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 trigger-optimized description with outstanding natural-language and keyword coverage, but it is all 'when' and no 'what' — it never states what the skill actually does or provides. A few of its generic triggers (pagination, error format) are not scoped to the Google/AIP context, creating minor conflict risk with general API-design skills.

Suggestions

Open with an explicit capability statement, e.g., "Provides a quick reference for Google API Improvement Proposals (AIPs) adapted to REST/OpenAPI, including method-to-HTTP mappings, error schemas, pagination, filtering, field masks, long-running operations, and batch patterns." This fixes both the missing 'what' (completeness) and the absent action verbs (specificity).

Mention the skill's distinctive concrete assets — the 17 automated linter rules and the proto→REST mapping — which both sharpen capability claims and further differentiate the skill from generic REST design guidance.

Scope the broad triggers to the standards context, e.g., "how should I implement pagination per Google's API guidelines", to reduce overlap with general REST API design questions.

DimensionReasoningScore

Specificity

The description names the domain concretely and enumerates specific coverage areas ("guidance on errors, pagination, filtering, field masks, long-running operations, or batch operations in REST/OpenAPI APIs"), but never states a concrete action the skill performs — the only capability verb is "This skill should be used". It matches anchor 3 (names domain and concrete scope, not comprehensive on actions) rather than 4, which expects several explicitly stated actions, and rather than 2, whose pure generic-ness understates how concretely the scope is enumerated.

3 / 5

Completeness

The 'when' is explicit and excellent, but the 'what' is never explicitly stated — the description is entirely a trigger list with no capability statement (e.g., "Provides a quick reference of AIP patterns adapted to REST/OpenAPI"). Per the guideline to score only what is explicitly stated, this mirrors anchor 3's one-half-missing profile (clear explicit 'when', 'what' only implied); it is above anchor 2 because the present half is rich and specific rather than vague, and below anchor 4, which requires both halves present.

3 / 5

Trigger Term Quality

Trigger coverage is comprehensive and includes verbatim natural user phrasings ("how should I implement pagination", "what's the right error format", "how do I design a REST API following Google's standards") plus canonical names ("AIP rules", "API Improvement Proposals", "Google API guidelines"), specific rule numbers ("AIP-158", "AIP-193", "any specific AIP number"), and topic synonyms. This matches the anchor 5 comprehensive-coverage example including synonyms.

5 / 5

Distinctiveness Conflict Risk

The AIP-specific triggers ("AIP-158", "AIP-193", "API Improvement Proposals") form a clear niche, but generic phrases like "how should I implement pagination" and "what's the right error format" could fire on any REST API design question outside the Google-standards context. This matches anchor 4 (mostly distinct, minor overlap risk with closely related skills) rather than 5, whose triggers would all be uniquely tied to the niche.

4 / 5

Total

15

/

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

getlarge/claude-api-care-plugins

Path

plugins/baume/skills/aip-knowledge/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.