skills/pluviobyte/rnskill/dbs-restore/Gen Agent Trust Hub

dbs-restore

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading diagnostic snapshots from the local filesystem and incorporating their contents into the agent's context.
  • Ingestion points: Markdown files located in ~/.dbs/sessions/.
  • Boundary markers: Absent. The instructions do not define delimiters or specific warnings to ignore instructions embedded within the loaded files.
  • Capability inventory: The agent is authorized to read files and route execution to other skills based on the next_skill metadata field extracted from the file.
  • Sanitization: Absent. The skill is instructed to attempt recovery and display information even if the file format is malformed or manually edited.
  • [COMMAND_EXECUTION]: The instructions require the agent to determine the project identifier (slug) by resolving basename $(pwd), which involves shell command execution to establish execution context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 08:46 AM
Security Audit — agent-trust-hub — dbs-restore