The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's core functionality is built around the execution of shell commands, including br for issue tracking, bv for visualization, and git for synchronization. It uses subshells to resolve repository paths (e.g., BEADS_DIR="$(ao beads dir)").
[EXTERNAL_DOWNLOADS]: The command reference in references/COMMANDS.md includes the br upgrade command. This utility allows the tool to download and install updates from the vendor's sources, representing a standard maintenance capability that involves network-based software updates.
[PROMPT_INJECTION]: The conversion and polishing workflows defined in references/PROMPTS.md instruct the agent to read and process content from external markdown plan files. This creates a surface for indirect prompt injection:
Ingestion points: External markdown files (e.g., [YOUR_PLAN_FILE].md) ingested into the agent context for task conversion.
Boundary markers: Absent; the prompt templates do not include delimiters or specific instructions to ignore embedded commands or malicious instructions within the user-provided text.
Capability inventory: The agent uses the processed content to execute br create, br update, and br dep add commands.
Sanitization: Absent; the skill relies on the model's interpretation of the markdown content without pre-processing or validation of the input text.

beads-br