The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions direct the user or agent to clone a specific repository from a personal GitHub account.
[REMOTE_CODE_EXECUTION]: The installation process involves executing code from an unverified external source via uv run eda.py and pip install -e .. This represents a high-risk pattern as the source repository (sibilleb/AAP-Enterprise-MCP-Server) is not associated with a trusted organization or the skill author.
[PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions.
Ingestion points: Data is ingested through tools like list_activations, get_rulebook, and list_event_streams.
Boundary markers: There are no explicit instructions or delimiters to isolate untrusted content from the agent's control flow.
Capability inventory: The skill has the ability to modify the environment by enabling, disabling, and restarting EDA activations.
Sanitization: There is no evidence of content validation or sanitization for the retrieved EDA rulebooks or stream data.

aap-eda