bear-notes
Warn
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill metadata specifies the installation of the
grizzlyutility usinggo install github.com/tylerwince/grizzly/cmd/grizzly@latest. This downloads and compiles source code from a third-party repository not associated with a known trusted organization. - [COMMAND_EXECUTION]: The skill utilizes shell commands to interface with the Bear application via the
grizzlybinary, including the use of pipes and input redirections for note creation and modification. - [CREDENTIALS_UNSAFE]: The skill instructions direct the user to store a Bear API token in a configuration file at
~/.config/grizzly/token. The tool subsequently reads from this sensitive file path for authentication. - [PROMPT_INJECTION]: The skill facilitates the ingestion of external data from Bear notes into the agent's context, which presents a surface for indirect prompt injection.
- Ingestion points: Data is retrieved via
grizzly open-noteandgrizzly tagscommands. - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat note content as untrusted data.
- Capability inventory: The skill includes capabilities to create, append to, and search notes using shell command execution.
- Sanitization: The skill does not implement sanitization or validation of the content retrieved from Bear before it is processed by the agent.
Audit Metadata