skills/konglong87/superpm/pm-agile/Gen Agent Trust Hub

pm-agile

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The Preamble section executes a bash script (check-update.sh) located at a relative path (../../) outside the skill's own root directory. This constitutes the execution of code not provided within the audited skill file.
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform local file system host modifications, specifically creating document directories (mkdir -p).
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8). It ingests untrusted user input via the AskUserQuestion tool and interpolates it directly into markdown documentation via the Write tool. The absence of boundary markers or sanitization logic means malicious input could be used to influence the behavior of other agents that process the generated files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 06:20 AM
Security Audit — agent-trust-hub — pm-agile