The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the gooseworks CLI and a suite of mcp__gooseworks__* tools to interact with its backend. This includes operations for ad generation, brand management, and cost estimation.
[EXTERNAL_DOWNLOADS]: The skill uses the gooseworks fetch command to download ad-analytics recipes and scripts from the vendor-owned goose-skills repository on demand.
[REMOTE_CODE_EXECUTION]: The agent is instructed to save downloaded scripts to /tmp/gooseworks-scripts/ and execute them to perform complex diagnostics such as campaign performance analysis, lead quality auditing, and competitor intelligence.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection (Category 8).
Ingestion points: Data is ingested via get_brand_kit, which is populated by a research pass that fetches content from external URLs provided by the user (website_url). It also ingests external project data through remix_community_ad.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or sanitize potentially malicious instructions embedded in the crawled website content or community projects.
Capability inventory: The skill has the capability to perform network API calls (submit_remix_batch, update_brand_kit), execute shell commands via the CLI, and write files to the local disk (write_file).
Sanitization: No explicit sanitization or validation of the ingested external content is mentioned before it is used to steer ad generation or update the brand kit.

goose-ads