The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external information from markets, competitors, and public datasets, which could contain malicious instructions meant to influence the agent's output.
Ingestion points: SKILL.md identifies external sources such as market reports, competitor websites, public datasets, and investor dossiers.
Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded instructions within the processed data.
Capability inventory: No executable scripts are included in the skill; it relies on the platform's default tools (e.g., web search).
Sanitization: Absent; the skill does not provide instructions for sanitizing, escaping, or validating external content.
[NO_CODE]: The skill consists entirely of markdown instructions and YAML configuration metadata. It contains no Python scripts, JavaScript, shell commands, or other executable files, significantly limiting its direct attack surface.

market-research