The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external sources, such as market research, competitor analysis, and existing project codebases.
Ingestion points: The skill utilizes tools like Read, Grep, and research workflows to pull information from external assets and websites.
Boundary markers: There are no instructions for the agent to use delimiters or safety framing to ignore potentially malicious commands embedded in the documents it researches.
Capability inventory: The skill possesses Write permissions to the local filesystem, which allows the agent to output analyzed content into discovery briefs and research dossiers.
Sanitization: The instructions do not define any sanitization or validation steps to filter external content before the agent incorporates it into its final reports.
[COMMAND_EXECUTION]: The skill employs the Grep tool and references the use of Glob to explore and document existing project structures. This allows the agent to search and read content throughout the accessible filesystem based on user-provided queries.

bmad-discovery-research