The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user and interpolates it into a Markdown document intended for consumption by other AI agents. Malicious instructions placed in the subject name could potentially manipulate downstream agents.
Ingestion points: The subject and author CLI arguments in dossier_briefing_core.py are the primary points where untrusted data enters the context.
Boundary markers: The generated Markdown uses standard headers and comments (e.g., ), but lacks strong delimitation or instructions to downstream agents to ignore embedded commands within user-provided fields.
Capability inventory: The skill includes file-writing capabilities to the local disk via the -o argument in the Python script.
Sanitization: There is no evidence of sanitization, escaping, or validation of the input strings before they are written into the Markdown file.
[COMMAND_EXECUTION]: The skill's core logic allows for the creation and overwriting of files on the local filesystem based on user-supplied paths via the -o or --output arguments. While this is the intended function of a dossier generator, it provides a vector for local file system interaction that should be scoped to prevent access to sensitive directories.

dossier-briefing