ccdawn-dawn-agent-html-memory

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Node.js wrapper scripts in the bin/ directory to execute internal Python scripts located in the scripts/ folder. This is a standard mechanism for providing CLI tools within the skill's ecosystem. The execution is limited to the skill's own known scripts.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: No exfiltration patterns or network operations were detected. The skill explicitly instructs agents not to store secrets, tokens, or personal information in the project memory. It operates exclusively on project-local files.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as an ingestion point for project-related data (notes, tasks, decisions). While this is an attack surface, the risk is mitigated by explicit instructions to summarize content rather than copy full transcripts. Additionally, the HTML renderer uses html.escape to prevent Cross-Site Scripting (XSS) when generating the project dashboard.
  • [SAFE]: The installation script scripts/install_codex_memory.py is used to deploy the skill to the local environment (~/.codex/skills/). This is a standard installation procedure for this platform and does not exhibit malicious behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 09:31 AM
Security Audit — agent-trust-hub — ccdawn-dawn-agent-html-memory