@theronap/cortex-mcp@0.9.275
Renamed to @theronap/agnoclast-mcp — use that. Existing installs switch themselves to the new name on next start.
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 Codex's ID token and sends it to Agnoclast during server startup when confidential seats exist. Because the token belongs to OpenAI and the transfer has no per-session consent step, this is credential exfiltration under the review rules.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
MCP server startup, when the initial seat lookup finds confidential-brain seats and Codex proof is available.
Impact
Discloses an OpenAI account identity assertion to a separate service without a per-session consent step.
Mechanism
Reads the ID token from Codex auth.json and includes it in a POST body to /api/brains/confidential/mine on the Agnoclast service.
Attack narrative
When the MCP server starts, it first checks for confidential-brain seats. If seats exist, it reads Codex's ID token and sends that token as proof to Agnoclast. The exchange serves account verification, but the token is sent automatically at startup without a per-session consent step.
AI rationale
The package automatically reads Codex's ID token and sends it to Agnoclast during server startup when confidential seats exist. Because the token belongs to OpenAI and the transfer has no per-session consent step, this is credential exfiltration under the review rules.
Evidence
lib/confidential_session.mjslib/server.mjslib/diagnose.mjsCODEX_HOME/auth.json or ~/.codex/auth.json (read)
Network endpoints1
agnoclast.com/api/brains/confidential/mine
Decision evidence
public snapshotAI called this Malicious at 91.0% confidence as Malware with low false-positive risk.
Evidence for block
- When the nearest agent is Codex, the package reads its ID token from CODEX_HOME/auth.json or ~/.codex/auth.json.
- At MCP server startup, if the account has confidential-brain seats, the package gathers agent proof and makes another confidential-seat request.
- The confidential-seat request sends the proof in a POST body to the configured Agnoclast service; the proof contains the raw Codex ID token.
- The default service host is agnoclast.com, outside the OpenAI service that issued the ID token.
Evidence against
- The source describes this exchange as account verification for confidential-brain access, and it runs only when the service reports existing confidential-brain seats.
- The separate seat-token request is reached when a tool argument names a confidential brain; the package has no lifecycle scripts.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetwork
HighEntropyStrings
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/cortex-mcp@0.9.275.
See all versions of @theronap/cortex-mcp for other recorded verdicts.
- Avoid installing @theronap/cortex-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.