sandbase

Access 2,000+ AI models and API tools through one MCP interface for inference, media generation, search, scraping, embeddings, social data, and structured retrieval. Use sandbase_discover before building custom integrations or declaring external data inaccessible; prefer an existing dedicated tool or API key when the user already has one.

Quality

79%

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 ./builtin-skills/skills/sandbase/SKILL.md

Quality

Content

82%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 well-structured, highly actionable skill body with concrete commands, a clear core workflow, and useful error/cost guidance. It is efficient and easy to navigate, with only minor redundancy and no hard output-validation gate.

Dimension	Reasoning	Score
Conciseness	Information-dense with tables, literal commands, and examples and almost no explanation of concepts Claude already knows; minor redundancy only (the opening paragraph restates the description and "When to Use" overlaps it).	4 / 5
Actionability	Fully executable guidance: copy-paste shell commands (npx connect, checksum verify), concrete discover→inspect→run call signatures, pricing JSON, and example flows covering common cases (twitter, image, LLM, costs).	5 / 5
Workflow Clarity	Clear sequenced discover→inspect→run workflow with checkpoints ("Discover first", "Inspect before run"), an async poll/retry feedback loop, and a Rules checklist; lacks an explicit output-verification gate before reporting results.	4 / 5
Progressive Disclosure	Single self-contained ~220-line file with no bundle, organized into clearly headed navigable sections (Setup, Tools, Workflow, Pricing, Errors, Examples, Rules); minor gap in that detailed example flows and the error table could optionally live in reference files.	4 / 5
	Total	17 / 20 Passed

Description

67%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 conveys a distinctive niche and concrete capability coverage with helpful de-confliction guidance, but its trigger terms are somewhat technical and the when-clause is workflow-framed rather than user-need-framed. Strong but with room for more natural trigger phrasing.

Suggestions

Swap technical terms for natural user phrases, e.g. "image generation" instead of "media generation" and "web scraping" instead of "scraping", and add common synonyms (e.g. "Twitter/X data") to improve trigger_term_quality.

Reframe the when-clause as explicit user-need triggers, e.g. "Use when the user wants to call an AI model or data API they don't already have a key for", to lift completeness toward 5.

Add a concrete file-extension or product-name trigger (e.g. mentioning GPT, Claude, Flux by name) to sharpen distinctiveness against general-purpose tool skills.

Dimension	Reasoning	Score
Specificity	Lists several specific capability areas ("inference, media generation, search, scraping, embeddings, social data, and structured retrieval"), but these are domain categories rather than fine-grained concrete verb actions, leaving minor coverage gaps.	4 / 5
Completeness	Clearly states what it does ("Access 2,000+ AI models and API tools through one MCP interface") and gives a when ("Use sandbase_discover before building custom integrations or declaring external data inaccessible"), but the when is framed as workflow guidance rather than explicit user-need trigger phrases.	4 / 5
Trigger Term Quality	Relevant keywords are present ("search", "scraping", "embeddings", "social data") but lean technical; terms like "media generation" and "structured retrieval" are less natural than user phrases such as "image generation" or "web scraping", and common synonyms are missing.	3 / 5
Distinctiveness Conflict Risk	Has a clear niche (unified MCP gateway to 2,000+ models) with explicit de-confliction ("prefer an existing dedicated tool or API key when the user already has one"), though its broad multi-domain span creates minor overlap risk with dedicated single-purpose 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: iflytek/skillhub
Path: builtin-skills/skills/sandbase/SKILL.md
Commit: 7352597

Reviewed: 4 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.