The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's core functionality relies on executing the 'br' (Beads Rust) and 'bv' (Beads Viewer) binaries to manage local databases and perform issue tracking operations.
[EXTERNAL_DOWNLOADS]: The br upgrade command facilitates a self-update mechanism, which involves downloading and executing code from remote sources to update the local binary.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting and processing untrusted data (issue titles, descriptions, and comments) that is subsequently presented to the agent in structured formats.
Ingestion points: Untrusted data enters the agent's context through commands like br list --json, br ready --json, br search, and br show <id> --json which retrieve content from the issue database.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when presenting issue data to the agent.
Capability inventory: The agent has access to powerful tools including shell command execution (via 'br' and 'bv'), Git operations for synchronization, and file system modifications within the project directory.
Sanitization: There is no evidence of sanitization or filtering applied to issue content before it is processed by the agent.

br