sdo-project-setup
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to initialize project environments using file copy commands (
cp) and to execute a local metadata synchronization script (set_docx_props.py) using theuv runcommand. These operations are confined to the local project workspace. - [DATA_EXPOSURE]: The
set_docx_props.pyscript accesses theUSERNAMEenvironment variable to populate the 'Author' property in the generated.docxdocuments. It also reads project-specific metadata from thePLAN.mdfile to update document titles and custom properties. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface by ingesting data from
PLAN.mdto update document properties. - Ingestion points: Metadata values extracted from the
PLAN.mdfrontmatter (e.g., title, tdoc, meeting). - Boundary markers: No explicit safety delimiters are used for the interpolated metadata.
- Capability inventory: The skill can perform file system writes (modifying
.docxfiles) and execute shell commands. - Sanitization: The Python script uses rudimentary string splitting and regular expressions to parse metadata, which limits the risk of sophisticated injection attacks against the script's logic.
Audit Metadata