Convert OpenAPI specs into token-efficient TML and .min formats.
npx skills add https://github.com/maf404/tml --skill tml
Please help me install this Agent Skill. Skill: tml Source: https://github.com/maf404/tml/tree/main Command: npx skills add https://github.com/maf404/tml --skill tml
This Skill solves the problem of verbose API schemas wasting tokens and confusing AI agents. Traditional formats like OpenAPI JSON use excessive tokens and reduce tool selection accuracy, making it expensive and unreliable for agents to discover and call APIs.
Use the tml skill to convert an OpenAPI spec into a .min tool registry file for AI agents.
High-intent search queries and answers about installing and using this skill.
You can reduce OpenAPI token usage for LLM agents by compiling verbose JSON schemas into a token-efficient .min format. This conversion significantly cuts token consumption while maintaining 100% tool selection accuracy for agent registries.
AI agents fail to select correct REST API tools because traditional OpenAPI JSON formats use excessive tokens and confuse tool selection accuracy. Converting these verbose schemas into an ultra-compiled one-line-per-tool .min format enables instant parsing and reliable execution.
A token-efficient tool manifest format for AI agents is a structured .min index that compiles API definitions into a one-line-per-tool representation. This allows agents to instantly parse tool intents, auth declarations, and parameters without wasting context window tokens.
Yes, you need Node.js to convert OpenAPI specs for AI agent consumption using this tool. An Anthropic API key is also optionally required if you want to use the AI-assisted generation and benchmarking features during the schema compilation process.
Yes, you can edit human-readable tool definitions in the TML source format before compiling an agent registry. This allows you to define intent descriptions, auth declarations, and structured parameters before generating the final ultra-compiled .min index file.
The best way to compile API schemas for an agent registry is to convert OpenAPI specs into a .min tool format. This process reduces token usage from over 1000 tokens to under 250 while preserving 100% tool selection accuracy for reliable AI agent execution.
KamiJeongCaptures SPEC Kit token metrics and stores them as run, summary, and baseline files for workflows.
Trac-SystemsRoute live Clawfinger gateway call turns to an external WebSocket agent.
caobingshengAutomate end-to-end AI agent development and orchestration workflows.
AradotsoGenerate command-line interfaces from MCP servers, OpenAPI specs, or GraphQL endpoints.
