- Anthropic Messages Token Count API - defines
POST /v1/messages/count_tokens, accepted request fields, and theinput_tokensresult. - Anthropic token counting guide - states that token counting supports system prompts, tools, and extended thinking; counts may include Anthropic-added system optimization tokens; the result is an estimate; and the endpoint is free subject to a separate requests-per-minute limit.
- Anthropic Claude Agent SDK overview - identifies the Agent SDK as the programmable Claude Code agent loop.
- Anthropic Agent SDK TypeScript reference - defines the optional
tools,systemPrompt, andsettingSourcesquery options used to describe the captured configuration. - Anthropic Agent Skills in the SDK - documents progressive disclosure: skill metadata is discovered at startup and full skill content loads when triggered.
@anthropic-ai/claude-agent-sdk@0.3.168- exact SDK package measured by the capture script.- Piebald-AI/claude-code-system-prompts - comparison catalog; the exact inspected commit is recorded in
capture-provenance.json.
Interpret · Interpretation constraints
Primary sources
Which external claims does the interpretation rely on?
What this proves
These are the Anthropic references used to define the API behavior, SDK configuration, token-count semantics, and skill-loading model.
- Input
- Anthropic API and SDK documentation
- Output
- Interpretation boundaries
- File
provenance/sources.md