groq-inference

Ultra-fast LLM inference on custom LPU hardware. OpenAI-compatible API at api.groq.com. Lowest latency in the industry (500-1000+ tok/s). Supports chat completions, vision, audio (Whisper STT + TTS), tool calling, JSON mode, and streaming. Free tier available. Inference only — no training.

Quality

73%

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 ./backend/cli/skills/ml-inference/groq/SKILL.md

Quality

Content

65%

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

A highly actionable, executable API reference with concrete examples across every feature. It loses points for token bulk and for keeping all model/pricing/reference material inline rather than splitting it into one-level-deep bundle files.

Suggestions

Move the model selection tables, rate-limit tables, and per-feature deep examples into one-level-deep reference files (e.g. MODELS.md, EXAMPLES.md) and link them from SKILL.md to improve progressive disclosure and token budget.

Trim the opening paragraph that repeats the description, and consider dropping the Groq-vs-vendor comparison table unless it directly drives a usage decision.

Add an explicit end-to-end workflow (e.g. choose model → set credentials → call with retry → verify output) with validation checkpoints so the reference becomes a guided process.

Dimension	Reasoning	Score
Conciseness	The body is dense and mostly non-redundant reference material, but ~250 lines with the intro paragraph restating the description and an extra vendor comparison table could be tightened; not a 3 because of that bulk.	2 / 3
Actionability	Fully executable, copy-paste-ready examples cover chat, streaming, JSON mode, tool calling, vision, audio, TTS, OpenAI compatibility, and retry logic with concrete model IDs and commands.	3 / 3
Workflow Clarity	Sections are clearly sequenced (credential check → quick start → features) and the retry helper plus credential verification give partial feedback loops, but there is no end-to-end multi-step workflow with explicit validation checkpoints for fragile operations.	2 / 3
Progressive Disclosure	No bundle files exist and the entire reference — model/pricing tables, all examples, troubleshooting — lives inline in one ~250-line SKILL.md; well-sectioned but content that should be split into reference files is inline, so it is not a 3.	2 / 3
	Total	9 / 12 Passed

Description

82%

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 specific, well-keyworded description that clearly names Groq's capabilities and distinguishes it from general inference providers. Its main gap is the absence of an explicit "Use when…" trigger clause, which caps completeness.

Suggestions

Append an explicit trigger clause, e.g. "Use when you need ultra-low-latency LLM inference, OpenAI-compatible API calls to open-weight models, or fast Whisper transcription/vision."

Consider trimming the marketing claim "Lowest latency in the industry" which is an over-claim that adds no trigger value.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — "chat completions, vision, audio (Whisper STT + TTS), tool calling, JSON mode, and streaming" — matching the top anchor for specific concrete actions.	3 / 3
Completeness	It clearly states what the skill does but lacks an explicit "Use when…" trigger clause; per the guidelines a missing explicit trigger guidance caps completeness at 2, even though "Inference only — no training" hints at a negative trigger.	2 / 3
Trigger Term Quality	Natural terms a user would say are well covered: "LLM inference", "OpenAI-compatible API", "Whisper", "vision", "JSON mode", "streaming", "free tier".	3 / 3
Distinctiveness Conflict Risk	"Groq", "custom LPU hardware", and "api.groq.com" carve a clear niche with distinct triggers unlikely to collide with other skills.	3 / 3
	Total	11 / 12 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: synthetic-sciences/openscience
Path: backend/cli/skills/ml-inference/groq/SKILL.md
Commit: 52845c3

Reviewed: about 11 hours 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.