antseed-connect

Connect coding agents, AI SDKs, and LLM tools to the AntSeed buyer proxy. Use when configuring Claude Code, Codex, OpenCode, Pi, OpenClaw, Hermes, GenLayer Studio, Vercel AI SDK, LangChain, or raw HTTP to route inference through AntSeed at localhost:8377.

Quality

83%

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

Quality

Content

71%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.

Highly actionable integration guidance with strong executable examples and a well-sequenced, verified setup workflow. Its main weaknesses are a monolithic single-file structure with no progressive disclosure and substantial duplicated boilerplate across the per-tool sections.

Suggestions

Consolidate the repeated 'How X talks to AntSeed' block and the openai-responses streaming caveat into a single shared 'Wire formats & protocol translation' section referenced once, instead of duplicating it under every tool.

Split the per-tool integration guides into separate reference files (e.g. references/claude-code.md, references/codex.md) and keep SKILL.md as an overview with one-level-deep links, since the file is ~1370 lines with no bundle files.

Trim explanatory passages that restate what each tool is (e.g. 'OpenCode is an MIT-licensed terminal coding agent ...') where the detail is not needed to perform the integration.

Dimension	Reasoning	Score
Conciseness	Per-tool config blocks are concrete and necessary, but the 'How X talks to AntSeed' section and the openai-responses streaming caveat are near-verbatim duplicates repeated across ~8 tools — padded sections that could be consolidated into a shared reference.	3 / 5
Actionability	Provides fully executable, copy-paste-ready config files (TOML, JSON, YAML, Python, TypeScript), CLI commands, curl examples, and install steps covering the common integration cases.	5 / 5
Workflow Clarity	A numbered 8-step universal setup with an explicit verify step (curl /v1/models), per-tool 'Test it' verification (buyer metering/status), and an 8-step canonical agent flow with troubleshooting feedback loops; minor validation gaps keep it just below 5.	4 / 5
Progressive Disclosure	No bundle files exist and all 1370 lines are inlined in one monolithic SKILL.md with no one-level-deep references; in-file section structure and navigation cues are good, but per-tool guides that could be split into separate files are inlined.	3 / 5
	Total	15 / 20 Passed

Description

95%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 tight, third-person description that clearly states both the capability and the trigger conditions, naming the specific tools it applies to. It is highly specific to the AntSeed niche with negligible conflict risk.

Dimension	Reasoning	Score
Specificity	Names concrete actions ('Connect ... to the AntSeed buyer proxy', 'route inference through AntSeed at localhost:8377') plus an enumerated set of real tools, but only two distinct actions with minor coverage gaps.	4 / 5
Completeness	Explicitly states what the skill does (connect tools to the buyer proxy / route inference) and an explicit 'Use when configuring ...' trigger clause with concrete trigger phrases.	5 / 5
Trigger Term Quality	Enumerates the exact tool names a user would say when configuring them (Claude Code, Codex, OpenCode, Pi, OpenClaw, Hermes, GenLayer Studio, Vercel AI SDK, LangChain, raw HTTP) — comprehensive natural-term coverage.	5 / 5
Distinctiveness Conflict Risk	Targets a specific niche (AntSeed buyer proxy at localhost:8377) with distinct, named tool triggers, giving it minimal overlap with other skills.	5 / 5
	Total	19 / 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 (1373 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: internet-court/internet-court-skill
Path: vendored/antseed/antseed-connect/SKILL.md
Commit: fa89195

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.