Tessl • The package manager for agent skills

SkillPublic

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

78%

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

77%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 highly actionable, well-organized reference with excellent executable examples and error-recovery guidance. Its weaknesses are length and staleness risk from inlined pricing/model tables, and the lack of any progressive disclosure — everything lives in one ~350-line SKILL.md with no bundle files.

Suggestions

Split the model catalog, pricing, and rate-limit tables into a references/ file (e.g. models.md) and keep SKILL.md as an overview with clearly signaled one-level-deep links.

Move volatile pricing figures into the reference file and note that they should be re-verified against https://console.groq.com/docs/models, reducing staleness risk in the main body.

Trim redundant content such as the vendor comparison table and the near-duplicate credential instructions (Credential Setup vs. Quick Start both cover GROQ_API_KEY).

Dimension	Reasoning	Score
Conciseness	The prose itself is lean with no over-explanation of concepts Claude already knows, but the ~350-line body inlines bulky, time-sensitive material — full model catalog tables with per-token pricing and speed figures — that will rot and could be tightened or split out, rather than being isolated in a deprecated/versioned section.	3 / 5
Actionability	Every section provides fully executable, copy-paste-ready code or commands: a credential check, quick start, streaming, JSON mode, a complete tool-calling round trip including the tool-result follow-up, vision, audio, and an exponential-backoff retry helper, plus a concrete troubleshooting table for common errors.	5 / 5
Workflow Clarity	For a simple single-purpose skill (call the Groq API), the flow is unambiguous — verify credentials, pick a model via the selection guide, call the API — and it includes explicit feedback loops: a retry function with exponential backoff for 429s and a Common Issues table mapping each error to a fix.	5 / 5
Progressive Disclosure	Sections are clearly organized with headers and navigable structure, but the body is a single monolithic file with no references: content that clearly belongs in separate files — the full model catalog, pricing/rate-limit tables, and troubleshooting — is inlined in SKILL.md rather than split into one-level-deep reference files.	3 / 5
	Total	16 / 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.

A concrete, information-dense description that clearly communicates capabilities and boundaries in third-person voice. Its main weaknesses are the absence of an explicit "Use when..." trigger clause and missing natural synonyms for the audio capabilities.

Suggestions

Add an explicit trigger clause such as "Use when the user asks for fast or low-latency LLM inference, needs an OpenAI-compatible endpoint with open models, or mentions Groq, LPU, or Whisper transcription."

Include natural synonyms users would actually say — e.g. "transcription" / "speech-to-text" / "text-to-speech" alongside "audio (Whisper STT + TTS)".

Trim the marketing claim "Lowest latency in the industry", which is an unverifiable over-claim, in favor of trigger-oriented phrasing.

Dimension	Reasoning	Score
Specificity	The description lists multiple specific concrete capabilities — "chat completions, vision, audio (Whisper STT + TTS), tool calling, JSON mode, and streaming" — plus concrete facts (API endpoint, speed range, free tier, inference-only boundary), giving comprehensive coverage in third-person voice.	5 / 5
Completeness	The "what" is clearly and concretely answered, but there is no "Use when..." clause or equivalent explicit trigger guidance — the closest is the implicit boundary "Inference only — no training" — so completeness is capped at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Good keyword coverage with natural terms like "LLM inference", "chat completions", "tool calling", "JSON mode", "streaming", and "free tier", but common user variations such as "transcription", "speech-to-text", or "text-to-speech" are only implied via "audio (Whisper STT + TTS)".	4 / 5
Distinctiveness Conflict Risk	The LPU hardware, speed, and "OpenAI-compatible API" framing carve out a clear niche that distinguishes it from general LLM skills, though it could still overlap with other inference-provider skills (Together, Fireworks) that share terms like "OpenAI-compatible" and "free tier".	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
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
Revision: 6a85742

Reviewed: 22 days ago

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.

groq-inference

Table of Contents

Is this your skill?