omni-inference

The core OpenAI-compatible inference endpoints: chat completions, embeddings, images, audio (TTS/STT), moderations, rerank, and the Responses API. The primary integration surface for AI agents.

Quality

51%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/omni-inference/SKILL.md

Quality

Content

32%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 custom capability guides are genuinely useful and executable, but they are buried under a massive auto-generated stub catalog that adds little actionable value. The skill is monolithic with weak progressive disclosure and no workflow structure.

Suggestions

Remove or collapse the ~90 stub endpoint entries that have no description beyond a title and an empty `curl -d '{}'`; keep only endpoints with real guidance, and link the full route list to the OpenAPI spec.

Split the per-capability guides (chat, image, TTS, STT, embeddings, search, fetch) into one-level-deep reference files and have SKILL.md serve as a concise overview that points to them.

Add a brief validated workflow (e.g. discover models → call endpoint → interpret documented error codes) with an explicit verification step, especially for batch/destructive endpoints.

Dimension	Reasoning	Score
Conciseness	Roughly 1,350 of 1,676 lines are auto-generated endpoint stubs like 'POST batches' followed by `curl -d '{}'` with no fields or explanation — heavily padded low-value content that crowds the context window.	2 / 5
Actionability	The custom section provides concrete, copy-paste-ready payloads for chat, image, TTS, STT, embeddings, search and fetch, but the large endpoint catalog offers only titles and empty `curl -d '{}'` bodies with no request fields or schemas.	3 / 5
Workflow Clarity	There is no sequenced multi-step workflow; the body is a flat endpoint reference. Destructive/batch endpoints (e.g. batches/{id}/cancel, batches/delete-completed) appear as bare stubs with no validation checkpoints or feedback loops.	2 / 5
Progressive Disclosure	It is a single monolithic file with no bundle files (references/scripts/assets absent); the bulk endpoint catalog is inlined rather than split into per-capability files, with only a one-line pointer to an external OpenAPI spec.	2 / 5
	Total	9 / 20 Passed

Description

71%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 clearly states what the skill covers with strong capability specificity, but lacks an explicit 'when to use' trigger clause. It is mostly distinct from sibling skills though trigger guidance is missing.

Suggestions

Add an explicit 'Use when...' clause naming concrete trigger phrases (e.g. 'Use when an agent needs chat completions, embeddings, TTS/STT, image generation, or rerank via an OpenAI-compatible gateway').

Include common synonyms and short forms users actually say (e.g. 'text-to-speech', 'speech-to-text', 'transcription') to broaden trigger-term coverage.

Tighten 'The primary integration surface for AI agents' into a more distinctive trigger phrase to reduce overlap with other omni-* skills.

Dimension	Reasoning	Score
Specificity	Enumerates many concrete capabilities — 'chat completions, embeddings, images, audio (TTS/STT), moderations, rerank, and the Responses API' — giving comprehensive coverage rather than vague abstractions.	5 / 5
Completeness	The 'what' is clearly stated (the endpoint surface), but there is no explicit 'Use when...' trigger clause, which caps completeness at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	Includes natural keywords a developer would say ('chat completions', 'embeddings', 'images', 'TTS/STT', 'Responses API') but misses common synonyms and file/extension variants, so coverage is good rather than comprehensive.	4 / 5
Distinctiveness Conflict Risk	'The core OpenAI-compatible inference endpoints' and 'primary integration surface for AI agents' carve a clear niche, with only minor overlap risk against sibling omni skills.	4 / 5
	Total	16 / 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
skill_md_line_count	SKILL.md is long (1677 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: diegosouzapw/OmniRoute
Path: skills/omni-inference/SKILL.md
Commit: c3945a7

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.