bear-notes

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the grizzly CLI tool from a third-party GitHub repository (github.com/tylerwince/grizzly) during the setup process.
  • [COMMAND_EXECUTION]: The skill uses shell commands to invoke the grizzly binary for creating, reading, and searching notes.
  • [CREDENTIALS_UNSAFE]: The skill references and uses a Bear API token stored at a fixed path in the user's home directory (~/.config/grizzly/token).
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it retrieves and processes content from Bear notes that could contain malicious instructions.
  • Ingestion points: Content is read into the agent's context via grizzly open-note, grizzly tags, and grizzly open-tag commands.
  • Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the notes being processed.
  • Capability inventory: The skill has the ability to write to the file system (modifying Bear's database via the CLI) and execute subprocesses.
  • Sanitization: No sanitization or content validation is performed on the data retrieved from Bear notes before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 10:44 PM
Security Audit — agent-trust-hub — bear-notes