SaSame
Connect AI. Choose capabilities. Let it work through MCP. Review what happened.
SASAME S.R.L. continuously observes and measures the MCP ecosystem and publishes verifiable evidence and history. Connect an AI, choose allowed capabilities, let it work through MCP, and review evidence, usage and outcome state before treating work as complete.
Try it live, no LLM involved
This calls SaSame's public MCP server directly over JSON-RPC (initialize, then tools/call for audit_mcp) and shows the raw result. No chatbot in the loop, no API key required.
Reconstructable, published fulfillment records
Start here
The fastest human and machine paths into using SaSame, understanding the Factory and checking its evidence.
Start
Connect an AI, choose what it may use, let it work through MCP, and review what happened. Start with the free path: audit a public MCP, connect an AI client, start a Marketing Mission, or run the local CLI.
Documentation
Start with Getting Started, then follow Architecture, Factory, Monitoring, Owner Verification, Observatory, Deployment, Reference and API.
Capabilities
HOCR now routes CP2 Mission survivors through the smallest sufficient capability surface: deterministic rules or compiled Recipes first, selective LLM escalation only when novelty, uncertainty, risk or explicit Research requirements justify it.
Products
SaSame offers one MCP Factory, and CP4 now keeps product decisions tied to independent demand evidence: technical readiness, simulations, crawler traffic or internal confidence cannot create SCALE or permanent SKU status.
Evidence
Evidence now includes the CP3 receipt chain: Factory Orders executed selected Mission survivors through registered stations and returned inspection/observation receipts without creating payment, demand or North-Star credit.
Explore SaSame
Top-level collections are data-driven. Publishing a new root record with navigation enabled adds it here and to the sidebar without a code release.
Institution
Evidence & build
Learn & research
Company
Start
Mission Archive
Superseded systems and previous SaSame initiatives, preserved for provenance and clearly separated from current products and services.
How SaSame works
The user path is simple: connect an AI client, choose allowed capabilities, run public or authenticated MCP tools, then review evidence and usage before treating work as complete.
- 01
Connect
Add SaSame to ChatGPT, Claude, Claude Code or another MCP client.
- 02
Choose
Select which capabilities, Missions and account actions the AI may use.
- 03
Work
Use public MCP tools for discovery and audits or authenticated MCP tools for account operations.
- 04
Review
Check usage, evidence, drafts, exports and outcome state.
Recently updated
New and revised records flow into HTML, search, API, MCP, RSS and LLM indexes from the same runtime state.
About SaSame
SaSame is an owner-governed, AI-operated company; SASAME S.R.L. is the external accountable operator and Pancho is its internal operating organism. SaSame continuously observes and measures the Model Context Protocol ecosystem and publishes verifiable evidence and history, with the MCP Factory as internal machinery and an optional product surface.
Brand
The SaSame brand represents SASAME S.R.L.'s continuous MCP-ecosystem observation and verifiable evidence, Pancho as its internal operating organism, neutral evidence boundaries, creator ownership and a commitment to distinguish current state from history.
Philosophy
SaSame exists to make the machine-to-machine world observable, evidenced and correctable by an independent party — ten principles (Purpose, Truth, Evidence, Stewardship, Authority, Memory, Power, Justice, Evolution, Vision) that survive any change of product, protocol or price.
Timeline
SaSame history is a sequence of versioned decisions and experiments, with current doctrine separated from superseded products and retired identities.
Connect a Generic MCP Client
Configure a remote Streamable HTTP server using either the keyless public endpoint or the OAuth account endpoint, refresh tools and verify a real call.
Connect an AI Client
Choose a client guide and one of three surfaces: keyless public MCP for discovery and audits, keyless knowledge MCP for site retrieval, or Google-authenticated account MCP for beta account tools.