writing-skills
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (MEDIUM): The file render-graphs.js utilizes child_process.execSync to run the system utility 'dot'.
- Evidence: execSync('dot -Tsvg', { input: dotContent, ... }) in render-graphs.js.
- Risk: The input data is extracted directly from the content of a markdown file. This pattern involves executing system tools on potentially untrusted content, which could lead to command injection or exploit vulnerabilities within the Graphviz parser itself.
- [PROMPT_INJECTION] (LOW): The persuasion-principles.md file advocates for the use of authoritative language and commitment framing to ensure AI compliance.
- Evidence: Encourages using imperatives like 'YOU MUST', 'Always', and 'No exceptions'.
- Risk: These patterns are identical to those used in adversarial prompt injection to override system constraints or bypass safety filters.
- [DATA_EXFILTRATION] (SAFE): No evidence of unauthorized data access, credential exposure, or network exfiltration was found.
Audit Metadata