The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted content from research goals and external summaries, which could contain instructions intended to manipulate the agent's reasoning.
Ingestion points: The skill reads technical goals, assumptions, and paper summaries from files including reaper-workspace/notes/clarified-goal.md, reaper-workspace/notes/paper-summary.md, and reaper-workspace/notes/literature.md.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the content of these files as data only or to ignore embedded instructions.
Capability inventory: The skill possesses file-writing capabilities, specifically writing to reaper-workspace/notes/problem-statement.md and reaper-workspace/notes/ideas.md.
Sanitization: The skill lacks sanitization, validation, or filtering of the external content before using it to generate the problem formalization.

formalize-problem