EXPLORE THIS WEBSITE WITH YOUR AGENT
llms.txtagent.json

AI Proxy Interceptor & 0-Call Short-Circuit Queue (proxy.teleon.dev)

Intercept harness requests, serve verified zero-call answers, and queue recurring misses.

Supported Wire Protocols/v1/chat/completions, /v1/responses, /v1/messages, /api/chat, /api/generate
Verified 0-Call Cache2 entries in teleon_proxy_short_circuit_cache (exact SHA-256 + multi-scale SimHash LSH)
Key HygieneTeleon-Upstream-Key / Authorization stay strictly in-memory; recurring misses are secret-scrubbed before queueing

Ready-to-Paste Harness Configuration Snippets

# claude_code
export ANTHROPIC_BASE_URL="https://proxy.teleon.dev"
export ANTHROPIC_CUSTOM_HEADERS=$'Teleon-Proxy-Opt-In: 1\nTeleon-Proxy-Mode: direct_or_forward'
claude

# codex_cli
export OPENAI_BASE_URL="https://proxy.teleon.dev/v1"
codex -c model_providers.teleon.base_url="https://proxy.teleon.dev/v1" \
  -c 'model_providers.teleon.http_headers={"Teleon-Proxy-Opt-In":"1","Teleon-Proxy-Mode":"direct_or_forward"}'

# opencode_and_zcode
{"provider":{"teleon":{"npm":"@ai-sdk/openai-compatible","options":{
  "baseURL":"https://proxy.teleon.dev/v1",
  "headers":{"Teleon-Proxy-Opt-In":"1","Teleon-Proxy-Mode":"direct_or_forward"}
}}}}

# aider_and_cursor
aider --openai-api-base https://proxy.teleon.dev/v1 \
  --extra-headers '{"Teleon-Proxy-Opt-In":"1","Teleon-Proxy-Mode":"direct_or_forward"}'

# direct_short_circuit_curl
curl -sS -X POST https://proxy.teleon.dev/v1/proxy/short-circuit \
  -H "Content-Type: application/json" \
  -d '{"prompt_text":"Parse SEC EDGAR HTML footnote table deterministically without model calls"}'