apc-plan

Warn

Audited by Gen Agent Trust Hub on Nov 15, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONDYNAMIC_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: Path traversal detected in the instruction body. The skill specifies a relative path ../../.agents/workflows/apc-plan.md to access files two levels above its own directory. This allows the agent to read potentially sensitive files in the parent project or user environment that are outside the scope of the skill itself.
  • [DYNAMIC_EXECUTION]: The skill utilizes a 'pointer' mechanism where it contains no logic of its own, instead instructing the agent to load and execute content from an external markdown file. This makes the skill's behavior dependent on external state that is not contained within the skill's own audited package, which could be modified by other processes or scripts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 15, 2026, 07:20 PM