The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves public TikTok post data and metadata, which is untrusted content that could contain instructions designed to influence the agent.
Ingestion points: Task results are fetched from the external API in scripts/tiktok_task.py using urllib.request.
Boundary markers: The script does not apply specific delimiters or instructions to the agent to ignore content within the retrieved JSON records.
Capability inventory: The script has the ability to perform network requests and write files to the local system.
Sanitization: While the script performs rigorous regex validation on input TikTok URLs, the resulting data retrieved from the API is processed in its raw format.
[DATA_EXFILTRATION]: The skill performs network operations to ai-factory.frevana.com to facilitate TikTok data collection. While this is the intended purpose, the domain is outside the standard whitelist.
[COMMAND_EXECUTION]: The skill's test suite in tests/test_tiktok_task.py uses subprocess.run to execute the local shell script wrapper. This is limited to the testing environment and targets the skill's own internal components.

tiktok-posts-collect-by-url