The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external specifications and ticket data to drive automated development tasks. This creates an attack surface where malicious instructions embedded in a specification or issue tracker could attempt to manipulate the agent's implementation logic or tool usage.
Ingestion points: SKILL.md instructs the agent to read specifications and tickets from an external issue tracker (referenced in Step 1).
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the processed data.
Capability inventory: The agent has the capability to write files, manage git branches, and execute other developer tools (such as tdd and code-review skills).
Sanitization: There is no evidence of content sanitization or validation of the input specifications before they are used to direct the implementer subagents.

implement-spec