The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external data from spec.md and the bd CLI tool's task descriptions, which constitutes an indirect prompt injection surface. Maliciously crafted task descriptions could attempt to influence the agent's behavior during code implementation or command execution.
Ingestion points: Processes task markers from spec.md and task details from bd ready command output.
Boundary markers: No explicit delimiters or boundary instructions are defined to separate untrusted task data from agent instructions.
Capability inventory: The skill possesses capabilities to write files (TDD implementation), execute shell commands (bd CLI), and perform version control operations (git commits).
Sanitization: There is no evidence of sanitization or validation of the data ingested from the Beads backend or markdown files.
[COMMAND_EXECUTION]: The skill performs shell command execution using the bd utility (e.g., bd ready, bd note) and version control tools to manage task state and record implementation notes.

flow-execution