aisa-twitter-api
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [DATA_EXFILTRATION]: The skill's scripts access local configuration and environment files to retrieve the required API key for authentication.
- The scripts
scripts/twitter_client.pyandscripts/twitter_oauth_client.pysearch for anAISA_API_KEYin~/.aisa/credentialsand various.envfile paths. - The retrieved secret is transmitted to the relay at
api.aisa.oneto authenticate requests. This behavior is documented and represents the primary authentication method for the skill. - [INDIRECT_PROMPT_INJECTION]: The skill fetches untrusted data from Twitter/X that could contain instructions intended to influence the agent's behavior.
- Ingestion points: Functions in
scripts/twitter_client.pyretrieve external text from tweets, search results, and profile data. - Boundary markers: The instructions do not define boundary markers or explicit safety guidelines to treat the ingested social media content as untrusted data.
- Capability inventory: The skill has the ability to post, reply, and quote on Twitter/X via
scripts/twitter_oauth_client.py. - Sanitization: No sanitization or content filtering is applied to the ingested tweet text before it is presented to the agent context.
Audit Metadata