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

Teleon Control Plane · /app/router

Splits multi-hour objectives into bounded ≤ 4,000-token steps for local or cloud models while enforcing strict Means vs. Ends human governance boundaries.

6-Core OverviewAtomic Queues (320)Persistent Stores (278)Muscle Memory & Fuzzy RecallFile VaultFeed StreamsExecution RouterOrientation

Mode A & Mode B Overnight Decomposition

Mode A (mode_a_server_steps): Server emits a verified step graph with per-step token budgets.
Mode B (mode_b_client_context): Server supplies a compact context pack and deterministic rubric gates so small client models decompose locally.

POST /v1/tenants/:tenantId/decompose/plan
POST /v1/tenants/:tenantId/decompose/runs/:runId/checkpoint

Means vs. Ends Boundary Guard

Autonomous adaptation is permitted for Means (lane, parser, prompt, model tier) inside policy. Any change touching Ends (permissions, connected systems, data scope, success criteria) requires human approval regardless of cost or latency savings.

POST /v1/tenants/:tenantId/governance/boundary-check