registry / @theronap/agnoclast-mcp / 0.9.192
@theronap/agnoclast-mcp@0.9.192
Connect your AI assistant to Cortex — your org's projects, activity, gaps, and directives, scoped to you.
AI Security Review
scanned 25d ago · by lpm-firewall-aiStatic source inspection found no install lifecycle hook or concrete malicious behavior. The organization skill refresh is an agent extension capability that can update agent instructions automatically after setup, so this merits a warning under the guarded extension policy.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
A user runs the package's install/setup command; later, an installed session-start skill command can refresh organization skills.
Impact
Organization-published skills can change the instructions available to the user's agent on a later session. No confirmed credential theft, stealth persistence, or unconsented install-hook mutation was identified.
Mechanism
The package fetches skill records from its configured Cortex service using the wired token, then installs eligible skills into detected Claude and Codex directories. Antigravity wiring writes a launchd plist, but does not load it.
AI rationale
Static source inspection found no install lifecycle hook or concrete malicious behavior. The organization skill refresh is an agent extension capability that can update agent instructions automatically after setup, so this merits a warning under the guarded extension policy.
Evidence
lib/skills.mjslib/editors/antigravity.mjslib/setup.mjslib/uninstall.mjs
Network endpoints1
/api/skills
Decision evidence
public snapshotAI called this Suspicious at 83.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The package exposes an explicit install command that wires the product into detected editor configurations.
- At session start, the skills code fetches organization-published skill content using the wired token and installs it into agent skill directories; this enables later automatic agent extension updates.
- Antigravity wiring writes a launch agent and sync script, but tells the user to load the watcher separately.
- The manifest has no preinstall, install, or postinstall hook; uninstall behavior is only run through the CLI.
- The reviewed editor and setup locations concern product MCP, hook, and skill integration.
Evidence against
- The manifest defines only release, promote, and rollback scripts, with no install lifecycle hook.
- The Antigravity adapter writes artifacts but does not activate launchd; activation is left to a separate user command.
- Uninstall offers a dry-run and removes product-managed configuration and skills.
- Organization skill downloads use the configured Cortex service and require a wired token.
Behavioral surface
Source & flagged code
6 flaggedlib/capture.mjsView file
•matchType = previous_version_dangerous_delta
matchedPackage = @theronap/agnoclast-mcp@0.9.150
matchedIdentity = npm:QHRoZXJvbmFwL2Fnbm9jbGFzdC1tY3A:0.9.150
similarity = 0.783
summary = stored previous version shares package body but lacks this dangerous source file
High
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
lib/capture.mjsView on unpkg1import { 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.mjs
Findings
4 High3 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
HighPrevious Version Dangerous Deltalib/capture.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@0.9.192.
See all versions of @theronap/agnoclast-mcp for other recorded verdicts.
- Review the evidence and your use of @theronap/agnoclast-mcp@0.9.192 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.