The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration attempts were detected. The skill uses standard Google Workspace integration tools for its intended purpose.
[NO_CODE]: The skill consists entirely of markdown instructions and metadata, without any executable scripts or binary code attached.
[INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data from external communication channels as part of its persona functionality.
Ingestion points: Gmail messages, Calendar entries, and Drive documents are read via the gws workflow commands (e.g., +standup-report, +email-to-task).
Boundary markers: There are no explicit boundary markers or delimiters defined in the instructions to isolate external data from the agent's logic.
Capability inventory: The agent possesses capabilities to write to Google Sheets and create Chat messages based on the information processed from external sources.
Sanitization: The instructions proactively recommend the use of a --sanitize flag for operations involving sensitive team data, which acts as a mitigation strategy.

persona-team-lead