The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external artifacts that are produced by other agent skills in the pipeline, which presents an attack surface for indirect prompt injection.
Ingestion points: The skill reads manifest.json and logic diagrams (02-logic-{component}.md) located in docs/reverse/{analysis}/. These files are inputs generated by previous stages of the reverse engineering process (SKILL.md Step 1).
Boundary markers: The instructions do not specify any delimiters or safety markers to differentiate between the data being verified and potential instructions embedded within the diagrams or markdown text.
Capability inventory: The skill has access to significant capabilities including Write (to create reports), Bash (for searching and counting flow branches), and various Read tools (SKILL.md frontmatter).
Sanitization: There are no explicit instructions for the agent to sanitize or escape the content ingested from the logic diagrams before using it to generate reports or executing verification steps.

re-verify-logic