Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Rebrand to Anvil AI. Remove CacheForge marketing copy. Normalize install commands.
What This Skill Does
Searches logs across Loki, Elasticsearch/OpenSearch, and AWS CloudWatch from a single interface. Natural language queries are translated to the appropriate query language (LogQL, ES DSL, or CloudWatch filter patterns) before execution. Read-only, it never modifies or deletes log data.
Instead of switching between Grafana, Kibana, and CloudWatch consoles and writing different query languages for each, you can search all configured backends with plain English from one place.
When to use it
- Triaging a production incident by finding errors across checkout and payment services
- Discovering which services threw exceptions during a specific time window
- Verifying a deployment did not introduce new error types
- Tailing live logs from a specific microservice
- Correlating errors between two dependent services to trace a root cause
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Find error logs from the checkout service in the last 30 minutes
- 1Checks which log backends are configured by running the backends script
- 2Translates the request into the appropriate query language for each configured backend
- 3Executes the search with a 30-minute time window and a default limit of 200 entries
- 4Groups results by error type, counts occurrences, and builds a chronological timeline
- 5Returns a structured summary with error patterns, root cause hypothesis, and recommended actions
Incident summary showing 47 log entries: 23 NullPointerExceptions starting at 14:02, 18 database connection timeouts from 14:05, root cause identified as an exhausted connection pool in the checkout service
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.