The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill aggregates data from external sources such as article titles and summaries into a final digest report, creating a potential vector for indirect prompt injection if the source data contains adversarial instructions. * Ingestion points: Signal data, summaries, and URLs are loaded in the workflow described in SKILL.md. * Capability inventory: The skill writes files to the local ops/trend-scout/ directory. * Boundary markers: While the skill uses a structured Markdown output template, it lacks explicit instructions to the agent to treat the ingested data as non-executable or to ignore instructions embedded within summaries. * Sanitization: There are no explicit data validation or sanitization steps mentioned for the external content before it is rendered into the final file.

write-weekly-digest