Security audit

Skill Architect

Security checks for vulnerabilities and agentic risk

Overview

Review is needed because an ordinary request to continue discussing a topic can display an unrelated saved conversation without confirmation.

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

Warning

Location: scripts/extract.py
Finding: In-place Skill modification bypasses mandatory approval and backup gates
Content: View full analysis
Impact

The original Skill's instructions are removed and replaced, and existing extraction outputs can be overwritten, before the required approval or recoverability checks. An unapproved or premature Agent invocation can therefore alter active Skill behavior without a verified recovery copy.

No suspicious patterns detected.