The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing standard database libraries including firebase-admin, @google-cloud/firestore, and dotenv via npm.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool for administrative tasks and includes a setup script (scripts/setup-firestore.sh) that enables Google Cloud APIs and creates security rule configurations.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to read and process untrusted data from various Firestore collections.
Ingestion points: The agent ingests data from Firestore collections such as users, agent_sessions, a2a_messages, and agent_memory during standard and agent-to-agent workflows.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat retrieved database content as untrusted data or to ignore embedded instructions.
Capability inventory: The agent has broad capabilities including Bash(cmd:*), Write, and Edit tools, which could be misused if malicious instructions are retrieved from the database and interpreted by the agent.
Sanitization: The provided instructions and code templates do not specify sanitization or validation logic for data retrieved from Firestore before it is processed.

firestore-operations-manager