The Agent Skills Directory

[DATA_EXFILTRATION]: The skill communicates with api.crawlora.net, which is the official API domain for the vendor crawlora-org. This is standard functionality for the skill's stated purpose of providing web intelligence utilities.
[CREDENTIALS_UNSAFE]: The skill correctly instructs users to manage their CRAWLORA_API_KEY using environment variables rather than hardcoding them, which is a recommended security practice.
[COMMAND_EXECUTION]: The helper script scripts/crawlora.sh uses curl to interface with the API. It employs set -euo pipefail and properly quotes variables (e.g., "${base}${path}") to prevent shell injection during execution.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data as it is designed to scrape and extract content from external URLs.
Ingestion points: External content is ingested via the /web/scrape and /extract endpoints defined in SKILL.md and reference/endpoints.md.
Boundary markers: The instructions do not currently include explicit boundary markers or directives for the agent to ignore instructions embedded within the scraped web content.
Capability inventory: The agent can perform further network requests via scripts/crawlora.sh based on the data it processes.
Sanitization: No specific sanitization or filtering of the scraped content is performed by the skill's scripts before the data is returned to the agent context.

web-utilities-research