Skip to content
SkillModel-led2,756 tokens in use

pinme-llm

Guides an agent to call OpenRouter-backed language models from a PinMe Worker, with streaming and web search, using the project API key.

3,763 stars on the repository MIT Licenseglitternetwork/pinme · skills/pinme-llm
npx skills add glitternetwork/pinme
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing pinme-llm costs

53tokens idle
2,756tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

networkdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter12/15
  • Structure11/25
  • Examples12/15
  • Output15/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 39

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Fair
  • Development activity24
  • Maintenance32
  • Community48
  • Maturity40
  • Momentum40

This skill documents two PinMe proxy endpoints for language models: a models list and chat completions in the OpenRouter format. The Worker authenticates with the PinMe project API key and project name, and never holds the real OpenRouter key.

It includes non-streaming and streaming TypeScript examples, a browser function that reads the server-sent event stream, the OpenRouter web search tool with result limits, an error table and helper functions that read PinMe and OpenRouter errors.

What You Get

  • A listModels function and a callLLM function for Workers
  • A streaming handler that passes the event stream through to the browser
  • A frontend reader that turns streamed chunks into text
  • A web search example using the openrouter:web_search tool with bounded results
  • An error extractor and a JSON helper for PinMe and OpenRouter responses

Common Use Cases

  • Adding a chat endpoint to a PinMe app
  • Streaming model output to a web page as it arrives
  • Letting a model search the web for a question, with a cap on results

What Makes It Unique

The Worker never sees a provider key: PinMe holds it and bills the project’s LLM balance, and a 403 means the balance is too low. The skill also states that PinMe has no raw search endpoint and web search goes through the model tool.

Under The Hood

The skill supplies code for a Worker that sends HTTPS requests to the PinMe platform with the project API key, which forwards them to OpenRouter and can search the web when the search tool is set. It does not run commands or scripts. The agent writes Worker and frontend code in your project, and each model call uses the project’s LLM balance.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers