1c-feature-dev

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the Bash tool to perform implementation tasks, which allows for the execution of arbitrary system commands. While intended for 1C development, this capability requires oversight to prevent misuse.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external codebase content to drive its logic.
  • Ingestion points: Phase 2 uses Grep, Glob, and metadatasearch to ingest content from the 1C codebase, while Phase 0 ingests user-provided arguments.
  • Boundary markers: The skill lacks explicit markers or instructions to treat analyzed code as passive data, increasing the risk that embedded instructions in the codebase could be followed by the agent.
  • Capability inventory: The skill possesses high-impact capabilities including file system modification (Write, Edit), system command execution (Bash), and the ability to delegate work to sub-agents (Task).
  • Sanitization: There is no evidence of sanitization or validation of the content retrieved from the codebase before it is incorporated into the architecture planning in Phase 4.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 08:33 AM