Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
v1.0.0: Official release
What This Skill Does
Streams agent actions, errors, and reasoning chains to a TiDB Zero cloud database in real-time as persistent audit logs. Survives container crashes and environment resets. Supports both user-supplied credentials and auto-provisioned temporary databases.
Unlike local log files, cloud-persisted logs survive container crashes and remain queryable after the agent environment is gone.
When to use it
- Retrieving the last 100 actions before a production agent crash
- Proving agent behavior for compliance audits
- Debugging a multi-step agent run that failed mid-execution
- Persisting logs across ephemeral CI/CD containers
- Tracking error patterns across repeated agent runs
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: What did the agent do before it crashed?
- 1Connect to TiDB Zero using TIDB_* credentials or auto-provisioned DSN
- 2Query the agent_logs table for the most recent entries
- 3Filter by ERROR level and sort by timestamp descending
- 4Return the last 20 log entries with message and metadata fields
A list of 20 timestamped log entries showing the agent's actions and errors leading up to the failure
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.