The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill accesses a sensitive Bear API token stored at ~/.config/grizzly/token for authentication. Exposure of this token allows programmatic access to the user's Bear notes.
[EXTERNAL_DOWNLOADS]: The skill's installation process downloads and installs the grizzly tool from github.com/tylerwince/grizzly, which is an unverified external source not included in the trusted vendors list.
[COMMAND_EXECUTION]: The skill uses the grizzly binary to execute various system commands for note creation and manipulation, introducing risks associated with command execution if inputs are not properly sanitized.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by reading data from Bear notes which could contain malicious instructions.
Ingestion points: grizzly open-note and grizzly open-tag commands in SKILL.md.
Boundary markers: No delimiters or warnings to ignore embedded instructions are implemented.
Capability inventory: File system access and command execution via the grizzly CLI tool.
Sanitization: No sanitization or validation logic is present to filter content retrieved from the notes.

bear-notes