The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill facilitates the execution of remote code through its installation instructions, which use npx -y @mvanhorn/printing-press install dataforseo --cli-only. This command downloads and runs a script from the npm registry to set up the local environment and install the binary tool.
[EXTERNAL_DOWNLOADS]: The skill downloads the dataforseo-pp-cli binary and its Node.js dependencies from the npm registry during the prerequisite setup phase.
[COMMAND_EXECUTION]: The skill's primary function is the execution of the dataforseo-pp-cli binary via shell commands. This tool manages local data in a SQLite database, performs API requests, and handles file system output.
[DATA_EXFILTRATION]: The skill exposes network egress capabilities that could be used for data exfiltration:
The --deliver webhook:<url> flag allows the agent to transmit command results to an arbitrary external URL via POST requests.
The feedback command allows sending locally stored feedback data to a remote endpoint if the DATAFORSEO_FEEDBACK_ENDPOINT environment variable is configured.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted data from external search engines and AI models.
Ingestion points: Data is ingested from DataForSEO API endpoints including Google SERP results, backlink audits, and responses from AI models like ChatGPT, Claude, and Gemini.
Boundary markers: The instructions lack specific guidance on using delimiters or ignore-instructions when the agent parses the output from these external sources.
Capability inventory: The skill has access to shell execution (dataforseo-pp-cli), file writing (--deliver file:), and network operations (--deliver webhook:).
Sanitization: There is no mention of sanitization, filtering, or validation of the data retrieved from external sources before it is returned to the agent's context.

pp-dataforseo