The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of external binaries from non-trusted sources using npx -y @mvanhorn/printing-press and go install github.com/mvanhorn/printing-press-library/.... These repositories are maintained by the skill author and are not part of the recognized trusted vendor list.
[DATA_EXFILTRATION]: The CLI tool implements a --deliver webhook:<url> flag, which enables the redirection of command output to any user-specified URL via HTTP POST. This capability can be exploited to exfiltrate processed data to external servers.
[COMMAND_EXECUTION]: The skill invokes a local binary eu-tenders-pp-cli and includes a sql subcommand capable of executing arbitrary SQL queries provided as strings, such as eu-tenders-pp-cli sql "SELECT...".
[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) due to its processing of external tender data.
Ingestion points: The skill ingests and processes EU public procurement notices from the TED database (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content within the processed notices.
Capability inventory: The skill possesses network capabilities (sync, webhooks), file system access (SQLite DB, feedback logs, profiles), and arbitrary command/SQL execution.
Sanitization: The instructions do not define sanitization or validation logic for data passed into the which natural language command or the sql execution command.

pp-eu-tenders