Back to skill

Security audit

Interagent Queue

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed ledger-to-log observer, but users should remember that its logs may preserve task and result details on disk.

Install this only if you want MIAB callback activity recorded locally. Keep secrets out of ledger task/result text where possible, review the configured log path, and periodically rotate or delete the log if it may contain sensitive work details.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill documentation describes capabilities that include environment-variable access, shell invocation, and file reads/writes, but it does not declare permissions or provide any explicit trust boundary. That creates a transparency and authorization gap: users or orchestrators may invoke a skill that can read state files and write persistent logs without realizing its effective privileges.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill is designed to monitor a transaction ledger and persist human-readable entries to a log file, but the documentation does not warn that potentially sensitive inter-agent transaction content will be stored on disk. Persistent logging increases the risk of unintended disclosure, retention of sensitive data, and later access by other processes or users with filesystem access.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The script persists formatted ledger entries directly to a log file, and those entries can include summarized task and result text from callback records. Because ledger content may contain sensitive prompts, results, identifiers, or operational details, this creates a real confidentiality risk if users are not clearly informed and if logs are retained with default filesystem access.

Session Persistence

Medium
Category
Rogue Agent
Content
PREREQUISITE: Requires the `miab-broker` skill to be installed and active.
It tails the append-only callback ledger (state/callbacks/ledger.jsonl) managed by miab-broker,
converts raw create / forward / return / resolve / cancel / fail events into human-readable log
entries using the agent identity map, advances a once-only cursor, and — when the live
toggle is on — writes the formatted batch to the log file ($CLAW_HOME/logs/interagent-queue.log).
Confidence
79% confidence
Finding
create / forward / return / resolve / cancel / fail events into human-readable log entries using the agent identity map, advances a once-only cursor, and — when the live toggle is on — writes the form

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.