Point it at a log file and get the incident, with line numbers, offline
Emerged finds new products worth a look, daily — what’s new today
Let a coding agent search, count and tail your logs through Kibana, read-only
Your agent can answer 'what errored in the last hour' from production logs without you opening Kibana or exposing Elasticsearch directly.
Kibana-agent is a Python command-line tool by Yevhenii Hyzyla that gives AI coding agents a way into Elasticsearch logs. It goes through Kibana's console proxy, so it works wherever you can already reach Kibana and needs no direct Elasticsearch access. Commands cover an index overview, search over a time window, document counts, live tail and date histograms, all returned as JSON, and every operation is read-only. It installs with uv and ships both as an agent skill for Claude Code, Cursor and Codex and as an MCP server for Claude Desktop and similar clients. MIT licensed, 29 commits, 8 stars.
Point it at a log file and get the incident, with line numbers, offline
Collapse millions of log lines into typed patterns before an LLM reads them
Inventory every AI agent on your machine and what it can reach
Self-hosted traces, logs and metrics that you and your agents can query
Webhook test URLs an agent can create and poll with two HTTP calls
A local ledger of every tool call your coding agents make