Warn
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the '@steipete/bird' package from the npm registry using the 'npm install -g' command to enable or update the CLI functionality.
- [COMMAND_EXECUTION]: The skill's primary operation revolves around executing the 'bird' CLI for various tasks such as reading tweets, searching, and managing account engagements.
- [DATA_EXFILTRATION]: The skill accesses highly sensitive authentication tokens (TWITTER_AUTH_TOKEN, TWITTER_CT0) and provides options to extract credentials from browser cookie databases using parameters like '--chrome-profile-dir'. While these are required for the tool's intended functionality, they represent access to sensitive user credentials.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection. 1. Ingestion points: Reads tweets, threads, and search results via 'bird read', 'bird home', and 'bird search' in SKILL.md. 2. Boundary markers: Absent; no delimiters are used to isolate external content. 3. Capability inventory: Includes the ability to post tweets ('bird tweet'), reply ('bird reply'), and follow users ('bird follow') in SKILL.md. 4. Sanitization: No evidence of filtering or validation of ingested content before it is processed by the agent.
Audit Metadata