pypi csoaistdioMITupdated 7d ago
Independent AI-governance measurement. We measure, we sign, we re-attest โ everyone can check.
What can you do with csoai?
Council of AI
Independent AI-governance measurement. We measure, we sign, we re-attest โ everyone can check.
Council of AI (CSOAI LTD, UK Companies House 16939677) is an independent measurement body for AI behaviour. We run AI systems on published, frozen instruments; issue the results as signed, offline-verifiable measurement credentials (Ed25519); and re-measure on a cadence so the evidence stays current. Measurement, not certification โ we do not certify, sell ratings, or remediate, and we take no money from anything we rank. Verification is free and loginless.
Live: councilof.ai on Cloudflare Pages (councilof-ai). Not Vercel.
Quick start
# Install via pip
pip install councilof_ai
# Or install via Smithery
npx -y @smithery/cli@latest install councilof-ai --client claude
What we measure
Live axis list and scores: GET https://councilof.ai/api/gspc (schema csoai.gspc-axes/0.5).
Snapshot 31 Aug 2026 from that API:
-
22 slots on the board ยท 15 measured ยท 7 UNMEASURED (declared empty, not a fail)
-
14 / 14 behavioural GSPC axes measured โ governance, safety, provenance, continuity, conformance, openness, machinery-conformity, care, cross-reality, detector-interop, art5-safeguard, swarm, affect, jail
-
Deterministic grading on frozen instruments โ no LLM-as-judge, no invented scores
-
UNMEASURED is first-class โ gaps are reported with their n and limits
-
335 signed cards (
n_cards == n_cells) ยท living stamp SIGNED (did:web:csoai.org#board-attestation-1) -
Methodology: doi:10.5281/zenodo.21991104
-
Overlay notes: ARC-AGI is an UNMEASURED overlay until a frozen gold bank exists; not a 23rd axis. Public GET stays 22ยท15ยท7. See
public/gspc-overlays.jsonandpublic/schema/gspc-axes-notes.json.
If this file and the API disagree, the API is right.
Verify
- https://councilof.ai/gspc-verify/ โ free, no login
- https://csoai.org/verify
- DID: https://csoai.org/.well-known/did.json
Hosting and deploy
| Host | Cloudflare Pages project | Deploy |
|---|---|---|
| councilof.ai / www | councilof-ai |
GitHub Actions deploy.yml โ wrangler pages deploy |
| csoai.org | csoai-site |
Wrangler (csoai-site-deploy.yml) |
Vercel is not the live host. The leftover Vercel Git links (csoai-v2-app, councilof-ai-src) were disconnected and those Vercel projects deleted on 31 Aug 2026.
Build (this repo):
npm run build:client
bash scripts/prerender-run.sh --dist dist/client --wait 900 --min 350
# GHA deploy.yml ships dist/client to Cloudflare Pages
Do not run npx vite build from the repo root (it picks up a dead src/). Do not vercel deploy this site.
Documentation
What we never do
- Certify AI systems or issue compliance badges
- Sell ratings, ranking position, or early sight of grades
- Remediate or recommend fixes in exchange for fees
- Take money in either direction from anything we rank
Surfaces
| Surface | Purpose |
|---|---|
| councilof.ai | Measurement body โ signed credentials, verify, scoreboard |
| csoai.org | Public site / DID apex |
| meok.ai | MEOK OS โ yours, on your keys |
License
MIT ยฉ CSOAI-ORG
Install
Add csoai to your client. Pick the one you use.
claude mcp add csoai -- uvx csoaicodex mcp add csoai -- uvx csoaiamp mcp add csoai -- uvx csoai{
"mcpServers": {
"csoai": {
"command": "uvx",
"args": [
"csoai"
]
}
}
}Add to `claude_desktop_config.json`, then restart Claude Desktop.
{
"mcpServers": {
"csoai": {
"command": "uvx",
"args": [
"csoai"
]
}
}
}Add to `~/.cursor/mcp.json`, or `.cursor/mcp.json` for a single project.
code --add-mcp '{"name":"csoai","command":"uvx","args":["csoai"]}'Or add the block manually to `.vscode/mcp.json` under `servers`.
{
"mcpServers": {
"csoai": {
"command": "uvx",
"args": [
"csoai"
]
}
}
}Add to `~/.codeium/windsurf/mcp_config.json`.
{
"mcpServers": {
"csoai": {
"command": "uvx",
"args": [
"csoai"
]
}
}
}Add to `cline_mcp_settings.json` via the MCP Servers panel.
{
"mcpServers": {
"csoai": {
"command": "uvx",
"args": [
"csoai"
]
}
}
}Add to `~/.gemini/settings.json`.
{
"mcpServers": {
"csoai": {
"type": "local",
"command": "uvx",
"args": [
"csoai"
],
"tools": [
"*"
]
}
}
}Add to `~/.copilot/mcp-config.json`, or run `/mcp add` inside the CLI.
{
"context_servers": {
"csoai": {
"command": {
"path": "uvx",
"args": [
"csoai"
]
}
}
}
}Add to your Zed `settings.json`.
uvx csoaiRun `goose configure`, choose **Add Extension โ Command-line Extension**, and paste this command.
Score
39 / 100
Incomplete
- Documentation16/25
- Maintenance19/25
- Trust13/20
- Capability0/15
- Install experience12/15
- Documents what it does and how to connect
- Has a resolvable package or endpoint
- Exposes at least one tool, prompt or resource
- README has substantive content
- Includes a code example
- Documents its configuration
- Mentions credentials or security posture
- Last commit 0 days ago
- Has a release history
- Repository is not archived
- Licensed MIT
- Namespace verified in the official MCP registry
- Claimed by its owner
- Published under an organisation
- 0 tool(s) documented
- Provides prompt templates
- Provides resources
- 12 documented install method(s)
- Published to a package registry
- Offers a hosted endpoint โ no local install
Version history
| Versions | Published |
|---|---|
| 0.1.2Latest | Aug 13, 2026 |