registry / @theronap/agnoclast-mcp / 0.9.236
@theronap/agnoclast-mcp@0.9.236
Connect your AI assistant to Cortex — your org's projects, activity, gaps, and directives, scoped to you.
AI Security Review
scanned 19d ago · by lpm-firewall-aiThe package has no npm lifecycle hook, and initial setup requires an explicit CLI command. However, setup enables automatic SessionStart synchronization of organization-published agent instructions, which creates a concrete agent extension lifecycle risk warranting a warning.
Static reason
No blocking static signals were detected.
Trigger
The user runs setup; afterward the installed Claude SessionStart hook runs skills repair at session start.
Impact
Updated skill instructions can change guidance and tool use in future agent sessions. The inspected source establishes this automatic update capability, but no malicious skill payload.
Mechanism
Skills repair uses the saved service token to fetch the account's skill list and writes returned skill content into managed Claude and Codex skill files without a separate per-update prompt.
AI rationale
The package has no npm lifecycle hook, and initial setup requires an explicit CLI command. However, setup enables automatic SessionStart synchronization of organization-published agent instructions, which creates a concrete agent extension lifecycle risk warranting a warning.
Evidence
bin/cortex-mcp.mjslib/editors/claude.mjslib/setup.mjslib/skills.mjs~/.claude/settings.json~/.claude/skills/<name>/SKILL.md~/.codex/skills/<name>/SKILL.md
Network endpoints1
agnoclast.com/api/skills
Decision evidence
public snapshotAI called this Suspicious at 89.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The manifest contains no npm install lifecycle hook; setup and installation are reached through explicit CLI subcommands.
- Explicit setup writes agent configuration and managed skills. It adds a Claude SessionStart hook that runs skills repair on later sessions.
- Skills repair automatically fetches the account's organization skill list using the saved service token when an agent directory is present.
- Returned skills are synchronized into managed agent skill files. Changed files are backed up and then replaced, allowing updated instructions to affect later sessions without a separate per-update prompt.
- Codex has no SessionStart hook, Cursor is capture-only, uninstall removes managed entries, and transcript lookup is limited to the requested session ID.
- Antigravity setup writes a launchd artifact but requires a separate manual load to activate it; the generated sync also depends on a local development checkout.
Evidence against
- The initial setup is explicitly invoked by the user, and the skill synchronization uses the configured Agnoclast service and existing account token.
- Changed managed skill files are backed up before replacement; the inspected source does not establish a malicious skill payload or an npm lifecycle attack.
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/agnoclast-mcp@0.9.236.
See all versions of @theronap/agnoclast-mcp for other recorded verdicts.
- Review the evidence and your use of @theronap/agnoclast-mcp@0.9.236 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.