skills/lev-os/agents/lev-align/Gen Agent Trust Hub

lev-align

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external sources that can be influenced by third parties. * Ingestion points: Reads project documentation (README.md, docs/), architectural decision records (ADRs), and dynamic repository data (git log, bd list issue tracker). * Boundary markers: There are no explicit markers or instructions to the LLM to ignore embedded commands within the ingested text. * Capability inventory: The agent has the ability to execute repository commands (git, bd) and perform file-write operations to update documentation. * Sanitization: No specific sanitization, validation, or escaping of the ingested data is specified before it is processed by the AI.
  • [COMMAND_EXECUTION]: Utilizes system commands like git log and bd list to retrieve project state information. These are standard operations for codebase analysis and are used within the local environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 09:43 AM