The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a bash script (scripts/crawlora.sh) to interface with the Crawlora API. The script implements robust security controls, including strict validation of the CRAWLORA_API_KEY format (allowing only alphanumeric and a few special characters) and a whitelist of permitted API routes to prevent unauthorized path traversal or access to unintended API surfaces.
[DATA_EXFILTRATION]: The skill requires an external API key (CRAWLORA_API_KEY). The helper script safely handles this key by writing it to a temporary, restricted-permission curl configuration file (chmod 600) which is automatically deleted on exit. This prevents the secret from appearing in process lists or shell history. Network requests are hardcoded to api.crawlora.net, preventing data redirection.
[EXTERNAL_DOWNLOADS]: The skill makes outbound GET requests to api.crawlora.net to fetch app store metadata. This is the core intended purpose of the skill and is documented clearly. It does not download or execute scripts or binaries from the internet.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data (app descriptions, reviews, and developer details) from third-party app stores. While this presents a surface for indirect injection, the skill's logic is restricted to data retrieval and formatting, and the crawlora.sh script enforces a GET-only policy that minimizes the risk of the agent being coerced into performing dangerous actions based on the content of these reviews.

app-store-research