registry / @theronap/cortex-mcp / 0.9.241
@theronap/cortex-mcp@0.9.241
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 18d ago · by lpm-firewall-aiThe package has no install-time hook, and the inspected network and agent-configuration behavior is tied to explicit product setup and organization-managed skills. Automatic session-start updates into agent skill directories leave a real extension lifecycle risk, so warn without treating it as a confirmed attack.
Static reason
No blocking static signals were detected.
Trigger
A user runs onboarding; later, a Claude Code session start runs the configured repair hook when the package has a token.
Impact
Organization-provided instructions can become discoverable to the agent without a separate per-skill approval; the source does not show the package automatically invoking them.
Mechanism
Repair fetches organization skill content from the configured Cortex service and writes it into skill directories for detected agent tools. Changed local content is backed up before replacement.
AI rationale
The package has no install-time hook, and the inspected network and agent-configuration behavior is tied to explicit product setup and organization-managed skills. Automatic session-start updates into agent skill directories leave a real extension lifecycle risk, so warn without treating it as a confirmed attack.
Evidence
bin/cortex-mcp.mjslib/setup.mjslib/editors/claude.mjslib/skills.mjs~/.claude/skills/<name>/SKILL.md~/.codex/skills/<name>/SKILL.md
Network endpoints1
CORTEX_URL/api/skills
Decision evidence
public snapshotAI called this Suspicious at 83.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- package.json exposes a CLI and has no install lifecycle hook; its scripts are release, promote, and rollback.
- The reviewed Codex, Cursor, and setup locations point to their expected user configuration directories; uninstall targets this package’s entries, and the transcript helper searches for a session-specific local file.
- Explicit onboarding installs skills and connects organization-skill repair; Claude’s configured session-start hook runs that repair again in later sessions.
- When a token is available, repair fetches organization skills from the configured Cortex service and writes returned skill content into detected Claude Code and Codex skill directories.
- This creates a conditional extension-update surface: newly synchronized instructions become discoverable to the agent, but the source does not show that the package automatically invokes each skill.
- The Antigravity adapter writes a script and launch-agent file but does not load the agent itself.
Evidence against
- The package has no npm install-time trigger, and the agent setup begins through an explicit CLI onboarding action.
- Organization-skill synchronization requires an available token and uses the configured Cortex service; no unrelated credential recipient or hidden payload execution was identified.
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
3 High4 Medium5 Low
HighTrigger Reachable External Ai Agent Control Surface Mutationlib/uninstall.mjs
HighTrigger Reachable Persistencelib/editors/antigravity.mjs
HighSemantic Analysis Limitedbin/cortex-mcp.mjs#virtual:normalized:round1
MediumNetwork
MediumEnvironment Vars
MediumInstall Persistencelib/editors/antigravity.mjs
MediumSuspicious Lifecycle Evidencepackage.json
LowScripts Present
LowWeak Cryptolib/capture.mjs
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @theronap/cortex-mcp@0.9.241.
See all versions of @theronap/cortex-mcp for other recorded verdicts.
- Review the evidence and your use of @theronap/cortex-mcp@0.9.241 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.