The Agent Skills Directory

[COMMAND_EXECUTION]: The skill frequently invokes shell commands, specifically npx goose-aeo@latest and node -e, to manage configurations, check environment variables, and perform analysis.
[EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the goose-aeo package from the NPM registry at runtime.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its data processing workflow:
Ingestion points: The skill ingests data from external websites during the audit command and processes responses from various third-party AI search engines (e.g., ChatGPT, Perplexity, Gemini).
Boundary markers: There are no explicit instructions or delimiters provided to the agent to distinguish between its own logic and potentially adversarial instructions embedded in the scraped content or AI responses.
Capability inventory: The agent has the capability to execute shell commands (npx), write to local files (.env, .goose-aeo.yml), and read configuration files.
Sanitization: No specific sanitization or filtering logic is described in the prompt instructions for handling untrusted external data.

aeo