platformBLOCK
Governed graph-native agent memory: knowledge extraction, fusion, hybrid RAG, scoped access tokens. A push for data sovereignty.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
The knowledge-infrastructure layer for enterprise AI.
Ground your data. Command your AI.
[](https://github.com/GCTRL-TECH/platform/actions/workflows/build-push.yml) [](./LICENSE) [](#-quick-start) [](#-access-control--compliance) [](#-quick-start)
GCTRL turns your scattered documents, databases, and code into one governed knowledge graph — then grounds your LLMs and agents on it, with enterprise-grade access control, entirely on your own infrastructure.
Most "AI memory" is a pile of vector chunks: fuzzy recall, no structure, no provenance, no permissions. GCTRL is graph-native — entities, typed relations, dossiers, and hybrid retrieval — and it runs 100% locally, so nothing leaves your building.
The pitch in one line: point GCTRL at your data, get a governed knowledge graph, and let any agent (Claude Code, Cursor, Codex...) read and write it as durable, access-controlled memory.
✨ Why GCTRL
- 🕸️ Graph-native, not a vector blob. Entities, typed relations, entity dossiers, and hybrid retrieval (dense + keyword + graph traversal) — with provenance on every fact. Ask "how does X connect to Y?" and get a real answer, not a fuzzy guess.
- 🔒 100% self-hosted, local inference. Runs on your hardware with local Ollama. DSGVO/GDPR by design — RAG sessions live in browser memor
408df86eb0fdOBSERVED · 2026-10-10Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add gctrl-mcp --env GCTRL_API_TOKEN=${GCTRL_API_TOKEN} -- npx -y [email protected]Exposed tools (6)
6 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
CRM | read | Customer Relationship Management |
Finance | read | Financial services |
GitHub | read | Connect GitHub repos, issues, PRs |
Healthcare | read | Healthcare and medical |
Legal | read | Legal and compliance |
Slack | read | Connect Slack workspace channels |
Trust audit
BLOCKgrade F · trust 45/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (12 observation(s))
- Network
- declared (22 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
/.env" ... curl
/.env" ... curl
/.env" ... curl
/.env" ... curl
".env" ... urllib.request
|| (o[0] == 169 && o[1] == 254) // link-local (incl. 169.254.169.254 metadata)
assert!(upstream_base_from(up, Some("http://169.254.169.254"), false).is_err());if host == "169.254.169.254" || host == "100.100.100.200" {assert!(validate_llm_base("openai", Some("http://169.254.169.254")).is_err());validate_llm_base("openai_compatible", Some("http://169.254.169.254")).is_err(),ssl: config.ssl ? { rejectUnauthorized: false } : false,ssl: config.ssl ? { rejectUnauthorized: false } : false,- DATABASE_URL=postgresql://GCTRL:GCTRL@postgres:5432/GCTRL
- PG_URL=postgresql://GCTRL:GCTRL@postgres:5432/GCTRL
- PG_URL=postgresql://GCTRL:GCTRL@postgres:5432/GCTRL
'postgresql://GCTRL:GCTRL@localhost:5433/GCTRL'
limes.jar
console.log(`[TokenManager] Refreshed token for connector ${connectorId}`);"Healthcheck": { "Test": ["CMD", "wget", "-q", "http://127.0.0.1/health"] }vault_url TEXT NOT NULL DEFAULT 'https://127.0.0.1:27124',
let api_base = format!("http://127.0.0.1:{}/api", state.cfg.port);let result = validate_runtime_input("ollama", Some("http://10.0.0.5:11434"));let result = validate_runtime_input("openai_compatible", Some("http://10.0.0.5:8080"));msg = "Error connecting to redis://default:pw@gctrl-redis:6379. Connection refused."
let secret = "ya29.super-secret-oauth-token-value";
Gates applied: no_behavioural_pass.
408df86eb0fdfull audit observations/trust-audit/mcp-server/gctrl-tech__platform.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-10 | 408df86eb0fd | BLOCK | F | 45 | first audit |
Questions
What is the platform MCP server?
Governed graph-native agent memory: knowledge extraction, fusion, hybrid RAG, scoped access tokens. A push for data sovereignty.
What tools does platform expose?
6 in total: 6 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is platform safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (45/100) and found 16 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does platform need?
It reads EMBEDDING_API_KEY, GCTRL_API_KEY, GCTRL_API_TOKEN, GCTRL_LICENSE_PUBLIC_KEY, GHCR_READ_TOKEN, GHCR_TOKEN, INTERNAL_API_SECRET, JWT_SECRET, LICENSE_HMAC_SECRET, LICENSE_PRIVATE_KEY_PATH, LICENSE_PUBLIC_KEY_PATH and NEO4J_PASSWORD from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does platform run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as gctrl-web at 0.1.0.
How current is this page?
The grade is for one exact copy of the source (408df86eb0fd), read on 2026-10-10. The repository is watched and re-audited when it changes.