gemini-history-analyzer
Warn
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill includes a file named '.security-scan-passed' that asserts the skill is safe, which is a metadata poisoning technique intended to mislead security analysis.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from Gemini conversation transcripts. Ingestion points: .txt files within the Takeout ZIP. Boundary markers: Absent. Capability inventory: Archive extraction (unar), keyword searching (grep), report generation, and file cleanup (rm). Sanitization: None.
- [COMMAND_EXECUTION]: The skill executes several shell commands including 'brew install', 'unar', 'find', 'du', 'grep', and 'rm' to perform its analysis tasks.
- [EXTERNAL_DOWNLOADS]: The skill installs the 'unar' utility via the well-known Homebrew package manager to handle archive extraction.
Audit Metadata