lev-social
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documents dependencies on the
postcrawlPython package and thebirdCLI tool. - [COMMAND_EXECUTION]: Executes shell commands using
birdfor Twitter searches andcurlfor interacting with the Exa search API. - [COMMAND_EXECUTION]: Integrates with a local
bdCLI tool for project tracking and management, which is consistent with the author's (lev-os) infrastructure. - [PROMPT_INJECTION]: The skill processes untrusted content from social media platforms, creating a potential surface for indirect prompt injection. (1) Ingestion points: Social media search results from
birdandpostcrawl. (2) Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the processing logic. (3) Capability inventory: The agent uses internal reasoning to synthesize data and thebdCLI to record findings. (4) Sanitization: There is no evidence of sanitization or filtering of the ingested social media content before it is processed by the agent.
Audit Metadata