story
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches update metadata and changelogs from the vendor's official GitHub repository (zenstory-ai/oh-story-claudecode) using tools like curl or the GitHub CLI.\n- [REMOTE_CODE_EXECUTION]: Includes an update utility that uses 'npx skills add' to download and install the latest skill version from the vendor's repository.\n- [COMMAND_EXECUTION]: Executes shell commands to start a local Node.js dashboard server, check tool versions, and open the system browser.\n- [DATA_EXFILTRATION]: Provides file management capabilities (read, write, delete) within a local workspace via the dashboard server. The server is secured by loopback-only listener binding and path resolution checks to prevent access outside the workspace.\n- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests content from local writer projects and analysis libraries.\n
- Ingestion points: Project files are read via the dashboard API or using file system utilities as described in SKILL.md.\n
- Boundary markers: The skill does not define explicit delimiters or instructions to ignore potential commands embedded within manuscripts or library files.\n
- Capability inventory: Actions include local file system modification, starting background Node.js processes, and executing version management commands.\n
- Sanitization: The dashboard server employs strict path validation and normalization to prevent directory traversal; the associated web interface sanitizes Markdown input.
Audit Metadata