Skip to content
Blopblop docs

Introduction

Blop is a local-first OpenTelemetry collector and command-line tool that lets you and your coding agent observe, investigate and verify the software you change.

Terminal window
blop up --detach                                              # start the collector on 127.0.0.1:27811
OTEL_EXPORTER_OTLP_ENDPOINT=http://127.0.0.1:27811 pnpm test  # run an instrumented workload
blop status --json                                            # check the collector and its store
blop audit --since 15m --json                                 # findings with trace IDs attached

The workload must already use an OpenTelemetry SDK: the endpoint variable alone sends nothing. Local collection, queries and audits need no account, API key or model. No instrumented app yet? The Quickstart records sample telemetry with blop demo.

What Blop gives you

Section titled “What Blop gives you”
  • Standard OpenTelemetry in. Any OTLP/HTTP exporter, in any language, sends traces and logs to the collector. There is no Blop SDK.
  • Local first. One process and one store on your machine, listening on 127.0.0.1. Local collection uploads nothing.
  • Output an agent can parse. Every query has --json output: one document on stdout, bounded results, documented exit codes and no prompts in JSON mode.
  • Evidence, not opinion. blop audit is deterministic and cites trace and span IDs. It reports what was observed together and never claims a cause from correlation alone.

Choose a Blop path

Section titled “Choose a Blop path”
  • Investigate: drill from collector status to traces, logs and audit findings.
  • Connect a coding agent: wire Claude Code, Codex, OpenCode or any MCP client to the collector.
  • Browser checks: write a .blop.ts file and run it with blop test to confirm a user flow works.
  • Blop Cloud: share projects, check runs and organization memory with your team.

To see how these fit together, read How it works.