The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data by scraping personal posts from X and LinkedIn. This creates a surface for indirect prompt injection where malicious instructions embedded in social media posts could attempt to influence the agent's analysis or output.
Ingestion points: Scraped content from Apify actors (apidojo/twitter-user-tweets-scraper and harvestapi/linkedin-profile-posts) and text provided by the user.
Boundary markers: The instructions do not define explicit delimiters or instructions for the agent to ignore potential commands embedded within the scraped social media content.
Capability inventory: The skill possesses file-write capabilities for configuration and output storage in the ~/.goose-skills/ directory and utilizes network-based scraping tools.
Sanitization: There is no evidence of content sanitization or filtering before the agent processes the scraped data for persona extraction and tone calibration.
[CREDENTIALS_UNSAFE]: The skill requires the APIFY_API_TOKEN environment variable to perform scraping tasks. It explicitly instructs the agent to prompt the user to provide this token if it is not found in the environment.
[EXTERNAL_DOWNLOADS]: The skill utilizes third-party Apify actors to fetch content from X and LinkedIn. While Apify is a well-known service, the skill's reliance on external actors for data retrieval is a core component of its workflow.
[COMMAND_EXECUTION]: The agent is instructed to modify the local file system by writing voice guides to ~/.goose-skills/voice-guides/ and updating the ~/.goose-skills/config.json file to register paths for sibling skills.

generate-voice-guide