Current company, Factory, documentation, research, evidence, product and API records are searched by default. Historical records are available only when you explicitly include them.
How does an AI agent use SaSame to accomplish a task, not just register an MCP?
Call resolve_and_run on the same public MCP endpoint with a goal, environment and constraints; it returns a bounded route and an honest execution state — only "verified" means something actually ran.
How is the SaSame Knowledge Platform structured?
One canonical Knowledge Database feeds the website, search, API, MCP, Markdown projections and future PDF outputs through versioned adapters.
How do HOCR and the Factory relate to 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.
How do I connect SaSame to another 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.
How do I connect SaSame to 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.
How do I connect the SaSame MCP to ChatGPT?
Create a custom MCP app using either the keyless public endpoint or the Google-authenticated account endpoint, scan tools, enable the app and verify a real call.
How do I connect the SaSame MCP to Claude?
Add either the keyless public URL or the Google-authenticated account URL as a remote custom connector, connect it, enable it and verify a real tool call.
How do I connect the SaSame MCP to Claude Code?
Register either the public or authenticated Streamable HTTP endpoint, confirm it in the MCP list, complete OAuth when required and verify a real tool call.
How are SaSame components deployed safely?
Deployments bind an immutable source SHA to a component release, health checks, external verification and a rollback target before traffic changes.
Where should I start with SaSame?
Start with Getting Started, then follow Architecture, Factory, Monitoring, Owner Verification, Observatory, Deployment, Reference and API.
How does the SaSame MCP Factory lifecycle work?
The Factory connects entry, blueprint, build, inspection, shipment, observation, repair and reverification through one evidence-backed lifecycle.
How do I begin using the SaSame MCP Factory?
Identify your MCP state, select the matching Factory entry point, then collect a reproducible baseline before changing anything.
How does BYO-LLM Guided Build actually work, step by step?
The creator's own LLM does the building from fixed instructions SaSame serves; SaSame runs zero LLM calls and deterministically verifies the artifacts the creator submits back — string/regex/length checks, never a fuzzy AI judgment.
How do documents enter the SaSame Knowledge Platform?
Documents enter through reviewed CMS edits, provider-neutral AI drafts or controlled migrations, then pass validation, CAS versioning, Markdown projection and receipt generation before public release.
Is there a command-line tool to check an MCP server’s readiness?
mcp-readiness is SaSame’s open-source command-line tool: run it with npx against any public MCP endpoint to get the same bounded readiness grade the Observatory publishes, with no account or key required.
What does Factory monitoring measure?
Monitoring records time-bounded reachability, protocol behavior, declared capabilities and selected compatibility checks with provenance and history.
How does the human Observatory experience relate to the evidence runtime?
The Knowledge Platform is the canonical human interface for the Observatory Factory station, while live-vps preserves signed JSON, feeds, badges, certificates, lookup and longitudinal evidence URLs.
What does owner verification prove?
Owner verification proves control of a defined MCP identity or related resource through a bounded control-proof method at a stated time.
Where are exact SaSame schemas and lifecycle states defined?
Reference pages expose versioned schemas, status enums, API contracts, evidence boundaries and release receipt fields used across human and machine interfaces.
When a shipped MCP breaks, what happens and who fixes it?
Observation detects drift and produces a scoped Repair Blueprint with hypotheses, not confirmed causes; the creator always applies the actual fix; SaSame never edits creator code; a successful fix is reverified under the same test contract and automatically reshipped.
How do I start using SaSame?
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.