The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the openfda-pp-cli binary using npx -y @mvanhorn/printing-press-library and go install github.com/mvanhorn/printing-press-library/.... These downloads originate from the author's own infrastructure and are used to provide the skill's primary functionality.
[DATA_EXFILTRATION]: The skill provides a --deliver webhook:<url> feature that enables the agent to POST command results to an arbitrary external URL. Additionally, the feedback command can be configured to send local data to an external endpoint via the OPENFDA_FEEDBACK_ENDPOINT environment variable. While functional, these features allow data to be sent to external, non-whitelisted destinations.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data which could contain malicious instructions.
Ingestion points: External FDA API responses (all search and list commands), user-provided notes in openfda-pp-cli feedback --stdin, and natural language strings processed by openfda-pp-cli which.
Boundary markers: The instructions do not define delimiters to isolate external data from the agent's system instructions.
Capability inventory: The skill has access to shell execution via Bash and the ability to perform network POST requests via the webhook delivery sink.
Sanitization: There is no mention of sanitizing or escaping the data retrieved from the FDA API or user feedback before it is displayed or delivered.
[COMMAND_EXECUTION]: The skill relies on the execution of the openfda-pp-cli binary with various subcommands and user-supplied arguments to perform its tasks.

pp-openfda