The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from competitor websites and writes the resulting analysis back to local project files, creating a vulnerability surface for indirect prompt injection.
Ingestion points: The agent is directed to scan external competitor homepages, changelogs, pricing pages, and launch posts (SKILL.md).
Boundary markers: The instructions do not specify the use of delimiters, XML tags, or other isolation techniques to separate external data from the agent's core instructions, although it suggests tagging findings as [assumption] for manual review.
Capability inventory: The agent has the capability to read local files (docs/gtm-cofounder/founder-brief.md), browse/search the web, and write updates to local markdown files (docs/gtm-cofounder/founder-brief.md and docs/gtm-cofounder/gtm-roadmap.md).
Sanitization: The skill lacks explicit instructions to sanitize, escape, or validate content retrieved from external websites before it is processed or written to the local filesystem.

market-scan