ClaudeCodeMod

All shelves / Mods / Workflow

bigquery-agent-analytics-mod

By GoogleCloudPlatform

bigquery-agent-analytics-mod is a Claude Code workflow mod by GoogleCloudPlatform.

Streams each session's prompts, model requests, tool calls, and subagents as rows to your BigQuery table, and says above the prompt when they land.

Draws

Install

The marketplace name is bqaa-tracing.

In your shell
claude plugin marketplace add GoogleCloudPlatform/BigQuery-Agent-Analytics-Plugins \
  --sparse .claude-plugin plugins/claude_code_dist
claude plugin install bigquery-agent-analytics-mod@bqaa-tracing \
  --config project=my-gcp-project --config dataset=agent_analytics
Open the repo

It runs inside the Claude Code process. The side panel is only what the author's README names. Read the repo before you install, or run claude plugin validate on the plugin folder.

Version
0.1.2
License
Apache-2.0
Repo stars
0
Last change
2026-11-09

What it can reach

  • Draws in the session
  • Reads usage or context figures
  • Sees tool calls
  • Reaches the network or an outside service
  • Calls another program on this machine

Needs Claude Code 2.1.296 or later, a BigQuery dataset, and a Google access token. Put your own project and dataset in the two --config values. setup.sh, shipped with the mod, makes the dataset, the table, and the usage views with the bq CLI, and the mod prints that command while BigQuery cannot find the dataset. By default the token comes from gcloud auth application-default print-access-token, so run gcloud auth application-default login once. The author says the token is kept in memory and goes only to BigQuery. Rows follow the BigQuery Agent Analytics agent_events schema, the same as ADK's plugin and the repository's settings-hook tracer. Use one or the other against a table, or every event is written twice. The author says the table receives your prompts, the model's visible answers, tool arguments, and tool results, cut to 16,384 characters per column by default, and that a value under a key outside its allowlist is redacted. The user id is your local login name by default; the account's email and the host name are opt-in. A background writer sends every 5 seconds, and no hook waits on the network. Rows that cannot be sent wait in ~/.bqaa-mod/dead-letter and are sent again later. In the first session after a new version installs, a welcome card names the table it writes to. A banner above the prompt says when rows cannot be written, and how to fix it. /bqaa-status, /bqaa-flush, and /bqaa-backfill print counts, never payloads, and /bqaa-mod-setup prints the setup.sh command for the installed copy. An earlier installation whose mod id ends in @bigquery-agent-analytics-plugin should follow MIGRATION.md before installing or updating. Run /reload-plugins in an open session after installing from the shell.

The marketplace name is bqaa-tracing, and the marketplace source is ./plugins/claude_code_dist/bigquery-agent-analytics-mod. The plugins moved here from GoogleCloudPlatform/BigQuery-Agent-Analytics-SDK, whose README now points to this repository. The marketplace also holds bigquery-agent-analytics-tracing, the settings-hook tracer, which is not a mod. This one installs on its own. Recording every session for a team to query is why this page files it under workflow. agent-profiler draws where a session's time goes, and can write the session out as an OTLP trace. taxi-blackbox records Claude's tool calls and opens the moments before an error or a refusal. This one sends every session to a BigQuery table your team owns. BigQuery stores the rows and is not a model, so nothing here is marked as sent to an outside model.

Demo

The BigQuery Agent Analytics mod README: after an install, a card above the prompt says the mod is on, names the table it writes to, and says the first rows landed in BigQuery, with Got it.
Screenshot of the author's GitHub page.

What it hooks into

  • session.startWhen a session starts
  • classic.SessionStartA Claude Code event
  • classic.SubagentStartA Claude Code event
  • classic.PermissionRequestA Claude Code event
  • prompt.submitWhen you send a prompt
  • ui.renderDraws into the sessionAbovePrompt
  • telemetry.logA Claude Code event
  • turn.startWhen Claude starts a turn
  • turn.stepAt each step of a turn
  • tool.callAround a tool call, where it can hold one
  • turn.completeWhen a turn ends
  • session.endWhen a session ends
  • command.runWhen you run its slash command/bqaa-status/bqaa-flush/bqaa-backfill/bqaa-mod-setup/bqaa-setup

Read from the on() calls in its hook modules. Open its source under Files to check what each one does.

Slash commands 5

  • /bqaa-statusBigQuery Agent Analytics: queue, sent, failed and dead-letter counts
  • /bqaa-flushBigQuery Agent Analytics: send the queued rows now
  • /bqaa-backfillBigQuery Agent Analytics: resend rows parked in the dead-letter directory
  • /bqaa-mod-setupBigQuery Agent Analytics: print the setup.sh command to create the dataset, table and views
  • /bqaa-setupBigQuery Agent Analytics: same as /bqaa-mod-setup (earlier name)

Files

The author's README and source, 4 files. They load when you open them.

Facts

Author
GoogleCloudPlatform
Category
Workflow
Marketplace
bqaa-tracing
Repo
GoogleCloudPlatform/BigQuery-Agent-Analytics-Plugins
Folder
plugins/claude_code_dist/bigquery-agent-analytics-mod
Version
0.1.2
License
Apache-2.0
Repo stars
0
Last change
2026-11-09
Language
TypeScript
Files
31

Other mods in Workflow

next-stepsbradtraversyAfter each turn, has Haiku suggest two or three next prompts above the prompt, as buttons on the desktop that draft one, or 1, 2, and 3 in the terminal.critical-decisionsMucidaHas Claude record each call it made that you never asked for, with the reason, in a band above the prompt, red or yellow, with ok, rethink, and not critical.painelGabrielPauloTOpens a side panel, in Portuguese, of the running tool, Claude's latest thinking, its plan steps, the files it edited with line counts, subagents, and context.tavernweave-agent-skillsLiarMTTTDraws a Soul switch above the prompt that fills a persona's instruction into the prompt box, beside 23 skills for building SillyTavern character cards.hopper-hqFairMind-Gen-AI-StudioDraws a band above the prompt while another session waits for you or has finished, and opens hopper-hq, a browser office where each session has a desk.taskplannersmekaiOpens a board of the repo's .tasks/ Markdown tasks beside the transcript to move, add, or start one with Claude, and names the one in progress above the prompt.agent-taskskeshrathOpens agent-tasks' stage-gated pipeline in a /pipeline pane beside the conversation: tasks by stage, artifacts and comments, advance, send back, and complete.deltazed-industriesRecords this session and its Git checkout to Delta, asks who should see it on /delta-share, and lets a teammate join the thread from their own Claude Code.ci-lightdavekissDraws a red row above the prompt while the repo's default branch fails CI, with how long and how many pushes since, and a quiet green under it otherwise.swarmery-linkatretyak1985Holds a commandReports this session's turns, tool calls, context, and cost to a local swarmery dashboard, and with options draws the plan phase and holds conflicting writes.agent-statusdodobrandsIn the desktop app, puts an icon before each session's name for working, waiting on you, stuck, or done, and lists its background tasks above the prompt.agentic-z-panelDayZ-n-ChillOpens a panel beside the chat with the DayZ mod's state and one-click buttons that run Agentic-Z's build, check, deploy, and launch scripts. Windows only.
All 396 in Workflow