p-blog-writer
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a helper script (
scripts/skillctx-resolve.py) designed to manage local configuration bindings. The instructions direct the agent or user to execute this script to resolve placeholders like{blog_drafts_dir}and{voice_profile_path}. The script's operations are confined to a specific local JSON file (~/.config/skillctx/config.json) for preference management. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data (notes, topic ideas, and audience descriptions) to generate content. While there is a potential surface for indirect prompt injection, the risk is mitigated by the 'Step-by-Step' workflow which requires explicit human approval for the outline and every section expansion before final assembly.
- [SAFE]: The skill operates as described, focusing on content creation and local configuration. It does not attempt to access sensitive system files (e.g., SSH keys), perform unauthorized network requests, or hide its logic through obfuscation.
Audit Metadata