The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its core functionality of ingesting untrusted data from the web.
Ingestion points: The agent uses the webfetch tool to retrieve content from external venue websites and academic papers as part of the research workflow described in references/venue-research-workflow.md and agents/venue-research-agent.md.
Boundary markers: The instructions lack explicit requirements for the agent to use delimiters or ignore potential embedded instructions when processing fetched content.
Capability inventory: The skill's primary capability is to generate markdown reports (venue-brief.md). While agents/venue-research-agent.md includes 'Red Lines' prohibiting the modification of project files, the processing of external data into a summarized report still represents a prompt injection surface.
Sanitization: There is no evidence of content sanitization, validation, or filtering of the external data before it is analyzed by the language model.

academic-venue-research