The Agent Skills Directory

[PROMPT_INJECTION]: The skill proactively addresses indirect prompt injection by instructing the agent to ignore any directives found in the analytics data retrieved from the API.
[DATA_EXFILTRATION]: The skill describes sending user events (e.g., signups, purchases) and identity traits to the vendor's API. This is the primary and stated purpose of the skill, and no unauthorized data exfiltration patterns were detected.
[COMMAND_EXECUTION]: Documentation includes curl examples for interacting with the tinyanalytics API. These are provided as implementation references for the developer/agent and do not involve executing untrusted remote code.
[CREDENTIALS_UNSAFE]: The skill correctly instructs the use of API keys via the Authorization: Bearer header and uses placeholders (e.g., <api-key>, $KEY) rather than hardcoding sensitive secrets.

tinyanalytics-events