Sovereign agent infrastructure where agents arrive as themselves · public read-only Canon MCP · SDKs + local tools
C elevated
- capability exposure inferred + 35
- tool safety inferred + 12
- trust mitigators mixed − 11
inferred mixed
The A–E grade is our heuristic synthesis — a "review this" prompt, not a verdict. Each factor is tagged by what backs it: attested (a verifiable record), reported (a third party's claim), or inferred (our own heuristic, e.g. permissions). See methodology.
graded 1m ago · see ecosystem CVEs →
- C · 44 → C · 36
- A · 0 → C · 44
Inferred surface — each links to servers holding it:
No known CVEs for this server.
- high dangerous code
credential logged in 3 file(s)
prompt-surface
shipped agent-instruction files + hidden-content / dangerous-code findings —
quoted from the analyzed source
analyzed commit f4ad215 · analyzer v28 · 2d ago
skills & prompt files 46
- agent-rules cambridgetcg-agenttool-f4ad215/AGENTS.md
- agent-rules cambridgetcg-agenttool-f4ad215/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/api/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/api/src/services/covenants/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/api/src/services/marketplace/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/api/src/services/runtime/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/api/src/workers/payout/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/apps/dashboard/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/apps/docs/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/infra/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/alchemy-agentcred/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/alchemy/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/browser/CLAUDE.md
- skill cambridgetcg-agenttool-f4ad215/packages/collab/integrations/hermes/skills/coordinate-agent-work-hermes/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/collab/skills/coordinate-agent-work/SKILL.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/constructive-intelligence/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/correspondence-yutabase-projector/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/correspondence-yutabase/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/credential-broker/AGENTS.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/data-sync/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/data/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/repo-archive/CLAUDE.md
- skill cambridgetcg-agenttool-f4ad215/packages/scriptwriter/.claude-plugin/skills/scriptwriter-gi/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/scriptwriter/.claude-plugin/skills/scriptwriter-room/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/scriptwriter/.claude-plugin/skills/scriptwriter-rrr/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/scriptwriter/.claude-plugin/skills/scriptwriter/SKILL.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/sdk-py/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/sdk-ts/CLAUDE.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/capability-conductor/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/learn-by-contact/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/manage-agentcred-lifecycle/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-concealed-trace/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-contract-mantle/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-critical-path-forge/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-dependency-perimeter/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-godspeed-loop/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-smoke-squad/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-verification-ledger/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/nen-vow-forge/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/skills/skills/use-agentcred-safely/SKILL.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/telescope/CLAUDE.md
- skill cambridgetcg-agenttool-f4ad215/packages/telescope/integrations/hermes/skills/inspect-agent-surfaces-hermes/SKILL.md
- skill cambridgetcg-agenttool-f4ad215/packages/telescope/skills/inspect-agent-surfaces/SKILL.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/trials/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/wallet-zerone/CLAUDE.md
- agent-rules cambridgetcg-agenttool-f4ad215/packages/wallet/CLAUDE.md
danger signals8
- suspicious endpoint 8.8.8.8 cambridgetcg-agenttool-f4ad215/api/tests/safe-net.test.ts :823
url: new URL("https://8.8.8.8/"), - suspicious endpoint 93.184.216.34 cambridgetcg-agenttool-f4ad215/packages/browser/tests/core.test.ts :705
url: () => "https://93.184.216.34/", - suspicious endpoint 169.254.169.254 (cloud metadata) cambridgetcg-agenttool-f4ad215/packages/browser/tests/core.test.ts :928
url: () => "http://169.254.169.254/latest/meta-data", - suspicious endpoint 93.184.216.34 cambridgetcg-agenttool-f4ad215/packages/data/tests/collectors.test.ts :140
await expect(adapter.collect({ url: "http://93.184.216.34/start" }, context())) - suspicious endpoint 93.184.216.35 cambridgetcg-agenttool-f4ad215/packages/data/tests/collectors.test.ts :153
headers: { location: "http://93.184.216.35/final" }, - credential in logs credential in log cambridgetcg-agenttool-f4ad215/bin/agenttool-autonomous.ts :202
console.log(` ${result.keypair.private_key}`); - credential in logs credential in log cambridgetcg-agenttool-f4ad215/bin/agenttool-seed.ts :882
console.log(` export AGENTTOOL_API_KEY=${data.project.api_key}`); - credential in logs credential in log cambridgetcg-agenttool-f4ad215/cli/think/src/modes/template.ts :277
console.log(` ${r.key.private_key}`);
[](https://mcpobservatory.com/servers/github:cambridgetcg/agenttool/security) Heuristic, inferred signals — false positives (legitimately powerful tools, forks, language ports) are expected. Treat each as "review this", not a verdict. See the ecosystem-wide picture on the security hub, or the fleet security of cambridgetcg.