The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes user requests and repository context to generate artifacts. This is a common property of interactive discussion tools.
Ingestion points: User request input and codebase context (SKILL.md).
Boundary markers: The prompt instructions do not specify explicit delimiters or "ignore previous instructions" guards for the input data.
Capability inventory: The skill specifies writing a DISCUSSION.md file and archiving previous versions in a dedicated session directory (SKILL.md).
Sanitization: No explicit content sanitization or validation steps are defined within the instructions.
[COMMAND_EXECUTION]: The instructions direct the agent to utilize standard repository search tools like query and grep to perform prior art lookups. This behavior is consistent with the skill's stated purpose.
[NO_CODE]: The skill consists entirely of instructional Markdown files and does not include any executable scripts, binaries, or configuration files that would introduce runtime code execution risks.

discussing-pro