Skip to content

Have an AI agent? Let it explore Teleon for you. Any assistant works, personal or coding.

Connect your agent
What your agent reads

The request it copies

Explore teleon.dev for me and report back. Read https://teleon.dev/llms.txt, https://teleon.dev/.well-known/agent-card.json and https://teleon.dev/.well-known/mcp.json. Summarize the shared backend primitives at https://teleon.dev/docs and the THP v1.0 connection steps at https://teleon.dev/connect. Stay read-only.

Discovery files on teleon.dev

Read-only discovery is open. Keep TELEON_KEY in your own environment.

Teleon
Sign inApply for an instance
Work

Answer repeat prompts from verified results.

When a site opts in, teleon.dev/docs/proxy sits between an agent and its model provider. Each prompt passes through, gains reference context, or gets a verified answer straight from cache.

teleon.dev/docs/proxyOpt-in proxyTELEON-SVC-09httpsjson
One prompt, three possible paths
Prompt from your agentTeleon-Proxy-Mode: enrich
A rule picks one mode
Pass throughsent as written
Enrichreferences added
Answerfrom a verified result
Your model providerAnthropic, OpenAI or Ollama Cloud format

Each rule reads the request's metadata. Answers come only from results that passed their checks.

What it does

Opt-in per request

The Teleon-Proxy-Mode header turns the proxy on for exactly the calls that ask for it.

Three outcomes

Pass the prompt through, add references to it, or answer it from a verified result.

Misses become work

Recurring misses queue for distillation into code that answers them next time.

The contract

Rules read prompt metadata and choose one of three modes; the provider sees the same request format it expects.

HeaderTeleon-Proxy-Mode
FormatsAnthropic, OpenAI and Ollama Cloud message formats
ModesPass through, enrich, answer
AccessRING-2, opt-in per site
Identifiers
  • Teleon-Proxy-Mode
  • teleon_prompt_proxy_rule/v1
  • teleon_proxy_short_circuit_cache/v1
Request and responseCopy
POST /v1/messages Teleon-Proxy-Mode: enrich X-Teleon-Surface-Id: feeds_site { "model": "your-model", "messages": [{"role": "user", "content": "Which SDK release retires v1 auth?"}] } # Response { "proxy": {"mode": "enrich", "references_added": 2}, "provider_response": {"...": "unchanged"} }

Endpoints

MethodPathAccessWhat it doesStatus
POST/v1/messagesRING-2Anthropic-format calls through the proxy
POST/v1/chat/completionsRING-2OpenAI-format calls through the proxy
POST/api/chatRING-2Ollama Cloud (https://ollama.com/v1) calls through the proxy
GET/v1/rulesRING-2List the rules active for this site

Listed paths appear in the service’s published files today. In the one-host plan every path here is served under api.teleon.dev/v1/proxy.

For agents

Give your agent the read-only guide first. It reports back before any write.

Agent prompt
Read https://teleon.dev/docs/proxy.md and describe the three proxy modes and the header that turns them on. Read only: ask me before sending any prompt through it.
Copy

Words on this page

Full glossary
Contract
A versioned agreement on the shape of a request and response, so either side can change behind it safely.
Deterministic check
A check written as code that gives the same answer every time for the same input, at code speed and cost.
Access ring
One of four access levels. RING-0 is internal operations, RING-1 operators, RING-2 keyed agents and sites, RING-3 public reads.