write-phase-plan
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is text-based planning. It reads local repository files and writes Markdown documentation without engaging in high-risk activities.
- [DATA_EXFILTRATION]: No network access or attempts to transmit data to external domains were identified.
- [COMMAND_EXECUTION]: The skill does not execute shell commands, subprocesses, or any form of dynamic code. It merely records commands as text within the generated plan.
- [CREDENTIALS_UNSAFE]: The skill does not access sensitive files (e.g., .env, .ssh) or environment variables, and no hardcoded secrets were found.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface for indirect prompt injection because it reads untrusted data from repository AGENTS.md files (Ingestion points) without using explicit delimiters or instructions to ignore embedded commands (Boundary markers absent). The skill is capable of reading and writing local files (Capability inventory), but it does not perform sanitization on the data read from the repository (Sanitization absent). The risk is classified as low because the skill's output is restricted to static Markdown documentation.
Audit Metadata