project-to-obsidian
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it reads and processes untrusted content from the target project directory.
- Ingestion points: Any file within the path provided to the
/p2ocommand (e.g., source code files, READMEs, config files). - Boundary markers: No specific delimiters or "ignore instructions" wrappers are used in the templates or described in the README to separate analyzed code from agent instructions.
- Capability inventory: The skill utilizes
Bash,Read, andWritetools, providing a significant surface for exploitation if an injection is successful (e.g., an attacker could place instructions in a code comment to delete files or write backdoors). - Sanitization: No evidence of input sanitization or validation of the content being read from the project files.
- [COMMAND_EXECUTION] (LOW): The skill requests the
Bashtool in its dependencies to perform project analysis and directory structure mapping. While a common requirement for project analysis tools, it grants the agent broad capabilities that could be abused if the agent is tricked by malicious project content.
Audit Metadata