Plug any agent into your team's Saya.
Before you guess, ask the team brain. Saya exposes a central remote MCP endpoint so external agents can query live, trust-graded team knowledge from the workspace Saya already lives in.
saya.computer ·
MCP: https://mcp.saya.computer/mcp
Add Saya as a Streamable HTTP MCP server at the live origin. When your client prompts, complete browser OAuth for your team; the per-team bearer belongs in the client credential store, never in this repo.
{
"mcpServers": {
"saya": {
"url": "https://mcp.saya.computer/mcp",
"transport": "streamable-http"
}
}
}Installs the secondary usage-guide skill for skill-aware agents. This teaches when to query Saya; the live MCP endpoint remains the product surface.
npx skills add creative-int/saya-pluginsAdd the marketplace, install the Saya plugin, then complete browser OAuth before the first team-brain query.
/plugin marketplace add creative-int/saya-plugins
/plugin install saya@sayaAdd this repo as a Codex plugin marketplace, install from /plugins, then authenticate the MCP connection for your workspace.
codex plugin marketplace add creative-int/saya-pluginsInstall Saya from the Cursor plugin marketplace, then authenticate the MCP connection for your workspace.
Cursor -> Settings -> Plugins -> Add marketplace -> creative-int/saya-pluginsTo preview the available skills without installing:
npx skills add creative-int/saya-plugins --list- Add the MCP server with the generated
.mcp.jsonblock above, or install the plugin through Claude Code, Codex, or Cursor. - When the client prompts for auth, complete browser OAuth with the workspace that owns the Saya brain. The bearer is per-team and should stay in the MCP client's credential store or environment, never in a tracked file.
- Start with
saya_statusif you need to confirm the live bridge, scopes, or workspace health. - Use
saya_contextbefore making team-specific assumptions. Ask focused questions like: "What has this team decided about the Saya MCP launch? Return verified decisions first and call out candidate or deprecated notes." - Use
saya_actonly for the curated approval-first memory save path. Treat it as a write that needs user intent, not a general automation channel.
If a client exposes raw MCP calls instead of a friendly tool picker, inspect
tools/list after OAuth and follow the current input schema advertised by the
server.
Saya exposes one central Streamable HTTP MCP server. Team identity and access come from per-team OAuth bearer auth plus workspace membership; never put tokens in repo files.
| Tool | Scope | Purpose |
|---|---|---|
saya_context |
saya.context |
Read bounded workspace context: provenance-graded team knowledge, lifecycle status, trust grade, decisions, skills, and captured workspace memory. |
saya_act |
saya.act |
Save approved team memory through Saya's curated, idempotent, approval-first action path. |
saya_status |
saya.status |
Check live MCP readiness, tool availability, auth posture, and Convex bridge health. |
The deployed endpoint is live on prod at
https://mcp.saya.computer/mcp. Production OAuth and
workspace membership are required; authenticated saya_context queries return
real, trust-graded team knowledge from Saya's Convex brain. saya_act is the
approval-first save_memory path, and saya_status reports live bridge health.
saya-team-brain— secondary usage guidance for agents deciding when to query Saya through MCP before answering. Useful for team memory, decision history, workspace norms, channel context, and "what does this team already know?" questions. Also answers toask-teamandteam-context.
The skill lives at
skills/saya-team-brain/SKILL.md.
See docs/team-brain-agent-tap.md for the
runner-side sample flow and proof boundary.
pnpm install
pnpm generate # regenerate all adapters from saya.config.ts
pnpm verify # drift check + typecheck + build/help + MCP smokepnpm smoke always checks unauthenticated protected-resource metadata. Set
SAYA_MCP_BEARER to additionally assert the authenticated tools/list surface.
MIT © creative-int
