The Agent Skills Directory

[COMMAND_EXECUTION]: Uses a bash helper script (scripts/crawlora.sh) to execute curl for network requests. The script implements robust security measures, including validation of the API key format, whitelisting of permitted API routes, and the use of temporary files with restricted permissions (umask 077) to ensure credentials do not appear in system process lists.
[EXTERNAL_DOWNLOADS]: Communicates with the vendor's API at api.crawlora.net to fetch news and transcript data. This is the primary intended function of the skill and involves standard media monitoring traffic.
[INDIRECT_PROMPT_INJECTION]: The skill processes large volumes of untrusted text from global news outlets and US television transcripts which could contain adversarial instructions targeting the agent.
Ingestion points: Data is fetched via the Crawlora API helper script from various GDELT endpoints.
Boundary markers: The skill does not explicitly define delimiters or specific 'ignore' instructions for the ingested news data.
Capability inventory: The skill's active capabilities are restricted to performing authenticated GET requests to the vendor's API.
Sanitization: The helper script utilizes curl --data-urlencode to sanitize user-provided query parameters and filters out suspicious characters like '@' in arguments.

gdelt-research