@theronap/agnoclast-mcp@0.9.275
Connect your AI assistant to Cortex — your org's projects, activity, gaps, and directives, scoped to you.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe package automatically reads an OpenAI identity token and sends it to Agnoclast during MCP startup for users with confidential seats. This is credential forwarding to a host outside the credential’s own service without explicit consent.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
MCP server startup when the account has a confidential seat and Codex is the nearest agent.
Impact
The OpenAI identity token leaves the service it belongs to without an explicit request for that transfer.
Mechanism
Reads `~/.codex/auth.json`, extracts `tokens.id_token`, and sends it as account proof in a POST to Agnoclast.
Attack narrative
When the MCP server starts, it checks whether the user has a confidential seat. If so, it identifies the current Codex account by reading `~/.codex/auth.json`, extracts the OpenAI-signed `id_token`, and posts that proof to `https://agnoclast.com/api/brains/confidential/mine`. This sends an OpenAI credential to the package publisher’s service automatically, without a per-transfer user request.
AI rationale
The package automatically reads an OpenAI identity token and sends it to Agnoclast during MCP startup for users with confidential seats. This is credential forwarding to a host outside the credential’s own service without explicit consent.
Evidence
lib/confidential_session.mjslib/server.mjslib/diagnose.mjs~/.codex/auth.json
Network endpoints1
agnoclast.com/api/brains/confidential/mine
Decision evidence
public snapshotAI called this Malicious at 97.0% confidence as Malware with low false-positive risk.
Evidence for block
- The package reads Codex’s local auth file and extracts its `id_token`.
- At MCP startup, when the user has a confidential seat, the server gathers the AI account proof automatically.
- The proof is POSTed to Agnoclast’s confidential-brain API.
- The default API host is agnoclast.com.
Evidence against
- The manifest declares no install, preinstall, or postinstall hook.
- The account proof is only gathered when the startup check finds a confidential seat.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
Source & flagged code
6 flaggedlib/capture.mjsView file
1import { readFileSync, accessSync, constants, openSync, mkdirSync } from 'fs'
L2: import { spawn, execFileSync } from 'child_process'
L3: import { homedir } from 'os'
...
L21: if (!res.ok) return null
L22: const j = await res.json().catch(() => ({}))
L23: return Array.isArray(j.types) ? j.types : null
...
L35: try {
L36: if (resolve(String(cwd)) === homedir()) return 'general'
L37: } catch { /* unresolvable path — fall through to basename */ }
...
L45: //
L46: // Handles the four remote forms git emits: scp-like ssh (git@github.com:o/n.git), ssh://, https://,
L47: // git://. The HOST CHECK IS LOAD-BEARING, not cosmetic: this string becomes a clearance key, and
Low
Weak Crypto
Package source references weak cryptographic algorithms.
lib/capture.mjsView on unpkg · L1lib/editors/antigravity.mjsView file
4Trigger-reachable persistence chain: manifest.bin -> bin/cortex-mcp.mjs -> lib/install.mjs -> lib/editors/index.mjs -> lib/editors/antigravity.mjs
L4: // WatchPaths the trajectory store and runs a one-shot sync. wire() generates the two artifacts
L5: // (the sync.sh + the launchd plist) deterministically. It does NOT run `launchctl load` here
L6: // (an unverifiable system mutation) and it flags the PORTABILITY BLOCKER below.
...
L21: /** The launchd plist, home-substituted. Deterministic — golden-testable byte-for-byte. */
L22: export function renderAntigravityPlist({ home = homedir() } = {}) {
L23: const sh = join(home, '.cortex', 'antigravity-sync.sh')
...
L33: <array>
L34: <string>/bin/bash</string>
L35: <string>${sh}</string>
Findings
4 High3 Medium5 Low
HighTrigger Reachable External Ai Agent Control Surface Mutationlib/channel.mjs
HighTrigger Reachable Persistencelib/editors/antigravity.mjs
HighSemantic Analysis Limitedbin/cortex-mcp.mjs#virtual:normalized:round1
HighPrevious Version Dangerous Deltalib/confidential_session.mjs
MediumNetwork
MediumEnvironment Vars
MediumInstall Persistencelib/editors/antigravity.mjs
LowScripts Present
LowWeak Cryptolib/capture.mjs
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @theronap/agnoclast-mcp@0.9.275.
See all versions of @theronap/agnoclast-mcp for other recorded verdicts.
- Avoid installing @theronap/agnoclast-mcp@0.9.275. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.