Back to skill

Security audit

Compliance Training

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for creating PersonWise compliance courses, but its update path can run a CLI/service-supplied command with too little local validation.

Review this before installing if you are comfortable authorizing PersonWise OAuth, uploading selected policy/training documents, consuming existing course credits, and installing a user-local CLI. Be especially careful with any proposed update action: approve only if it is clearly the expected PersonWise update and you trust the source.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:109
Finding

Execution of Remotely Supplied Update Commands Without Local Validation

Content
View full analysis
` with the directory of this installed Skill (the directory containing this Skill's SKILL.md). Never run `doctor` or a generic capability preflight to check freshness; the `update check` command above is the freshness check. Never ask more than once per component per session, and never substitute another command, flag, origin, or download path for the printed `action`. ``` ### Technical Analysis The Skill instructs the agent to execute an `action` command returned by the PersonWise CLI or its remote service. The command is treated as an opaque shell command: the Skill requires it to be run “exactly” and does not require the agent to parse its arguments, verify its executable and origin, reject shell metacharacters, or compare it against a local allowlist. User approval is required before execution, which reduces accidental installa ...[truncated 1966 chars]
Remediation
View remediation
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Scope Creep

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The bootstrap script performs an external download and installation workflow even though the skill metadata declares only Bash permission. That mismatch weakens users' and platform policy expectations around what the skill can do, and any network-capable installer increases supply-chain risk because it fetches executable content from a remote host and writes it into a user-local executable path.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · assets/bootstrap.sh (reported line 165)May include surrounding context.

sh
echo "Release executable checksum mismatch." >&2
  exit 6
fi
chmod 700 "$candidate"
if [ "$(uname -s)" = "Darwin" ] && [ "$native_signature_status" = "verified" ]; then
  codesign --verify --deep --strict --verbose=2 "$candidate"
  spctl --assess --type execute --verbose=2 "$candidate"