hugo-write-post
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs authorized analysis of local content files within a Hugo repository to extract writing style metrics. This behavior is transparently documented and serves the primary function of the skill without requesting excessive permissions.\n- [COMMAND_EXECUTION]: The skill instructions include the execution of a local utility script (
scripts/analyze_style.py) and the Hugo CLI (hugo new,hugo server). These are standard development tools and the script uses only Python standard libraries, posing no external dependency risks.\n- [PROMPT_INJECTION]: The skill processes existing blog posts from thecontent/directory to learn the author's voice, creating a potential surface for indirect prompt injection. \n - Ingestion points: Reads all Markdown files within the user-specified
content/directory.\n - Boundary markers: No explicit delimiters or instructions are used to separate the analyzed style data from the generation prompts.\n
- Capability inventory: The skill can execute local Python scripts, invoke the Hugo CLI, and write new Markdown files to the local repository.\n
- Sanitization: Content is read and parsed directly without sanitization of potential prompt injection sequences. However, since the data source is the user's own repository, the risk is considered low and consistent with standard RAG patterns.
Audit Metadata