The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from existing files during its 'UPDATE' and 'FIND' modes.
Ingestion points: The agent reads existing stakeholder briefs from user-specified paths or search locations such as communications/, briefs/, stakeholder/, and updates/.
Boundary markers: There are no explicit instructions or delimiters used to separate the content of the existing files from the agent's instructions, nor are there warnings to ignore embedded commands within those files.
Capability inventory: The skill has the capability to search the file system and write files to the communications/ directory. It does not appear to have network access or arbitrary command execution capabilities.
Sanitization: The skill lacks sanitization or validation of the content read from external files before processing it into a new brief.

stakeholder-brief