The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a bundled bash script (scripts/crawlora.sh) to interact with the Crawlora API. The script implements robust security controls:
Input Validation: It enforces strict alphanumeric format checks for the CRAWLORA_API_KEY to prevent command injection.
Path Whitelisting: The script uses a strict whitelist (case-match and anchored regex) to ensure only documented API routes for the chrome-extension-research skill can be accessed.
Secret Management: Instead of passing the API key as a command-line argument, it writes the key to a temporary file with restricted permissions (chmod 600) and uses curl's --config flag, ensuring the key is not visible to other users on the system via process monitoring tools.
Input Sanitization: Query parameters are URL-encoded using curl -G --data-urlencode, and the use of the @ symbol in arguments is prohibited to prevent unauthorized file inclusion attacks via curl.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to https://api.crawlora.net/api/v1. This is the official API endpoint for the skill's vendor and is used to retrieve extension metadata, history, and privacy disclosures as intended by the skill's primary purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (Chrome Web Store listings and reviews). While this is a common ingestion point for third-party content, the skill's design limits the risk by using a highly restricted helper script that lacks dangerous capabilities (such as writing to the filesystem or executing arbitrary commands) that could be exploited by malicious data.

chrome-extension-research