aikey-mvp
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes a bundled shell script (
scripts/top_posts.sh) to calibrate the AI's tone by analyzing and ranking the user's historical post performance from a local CSV database. - [COMMAND_EXECUTION]: Uses the system command
pbcopyto automatically place generated text into the user's clipboard. This is a core feature designed to facilitate the publishing workflow on external platforms. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, including chat screenshots and fragmented notes, which could theoretically contain malicious instructions. The skill effectively mitigates this surface by requiring the agent to summarize a "one-sentence core claim" and obtain user approval before generating the final output.
- [FILE_ACCESS]: Reads and writes to the local workspace (vault) to manage content drafts, performance pipelines, and historical memory. These operations are consistent with the skill's purpose as a content management assistant.
Audit Metadata