The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions involve reading sensitive API keys from a local JSON file at ~/.gooseworks/credentials.json to authenticate proxy requests.
[COMMAND_EXECUTION]: Employs python3 -c one-liners to parse local JSON configuration files and export values to the shell environment.
[EXTERNAL_DOWNLOADS]: Utilizes npx for installation (npx goose-skills install) and authentication (npx gooseworks login), which downloads and executes code from the npm registry.
[COMMAND_EXECUTION]: Performs network requests via curl to the api.gooseworks.ai infrastructure to perform SEO tasks.
[PROMPT_INJECTION]: The skill processes untrusted data from external websites (via SEO crawling), which presents an indirect prompt injection surface.
Ingestion points: Website content is retrieved from external URLs using Scrapegraph and Tavily APIs as defined in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when processing site content.
Capability inventory: The skill uses curl for network operations and python3 for local command execution.
Sanitization: No sanitization or filtering logic is provided for the ingested website data.

seo-analyzer