Claude Code usage telemetry and repository CI status are two separate signals: OpenTelemetry exports session, cost, token and tool-activity data, while GitHub Actions runs Claude workflows and reports checks on repository changes. Configure and verify each system independently; enabling one does not set up the other.
What do “Claude Code usage” and “CI status” mean?
Usage is operational telemetry from Claude Code sessions: metrics and events, with optional traces, that an organization can send to an observability backend. CI status is the result of a workflow triggered by repository activity, such as a push or pull request. Anthropic describes the telemetry system as a way to track Claude Code usage, costs and tool activity across an organization through OpenTelemetry (OTel). Anthropic’s monitoring guide and its GitHub Actions guide cover different configurations.
As an Amazon Associate I earn from qualifying purchases.
How do I set up Claude Code usage telemetry?
Enable telemetry explicitly, select the signal exporters you want, and configure the destination. For OTLP, the exporter protocol and endpoint must match your collector. This shell example follows Anthropic’s documented gRPC pattern; localhost:4317 is an example endpoint, not a universal production address.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallexport CLAUDE_CODE_ENABLE_TELEMETRY=1
export OTEL_METRICS_EXPORTER=otlp
export OTEL_LOGS_EXPORTER=otlp
export OTEL_EXPORTER_OTLP_PROTOCOL=grpc
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4317
claudeSet any authentication headers required by your collector as well. The exact variables and accepted values depend on the exporter and backend configuration; consult the current monitoring documentation before applying this example to a hosted service.
Choose an exporter that fits your observability stack
- OTLP: Send metrics and logs to an OpenTelemetry-compatible collector or backend. This suits teams that already operate an OTel pipeline.
- Prometheus: Use the documented Prometheus metrics exporter when it fits the organization’s metrics collection setup.
- Console: Use console output for debugging configuration locally, rather than treating it as a production monitoring destination.
Metrics and events can be useful without exporting traces. Select the signals your monitoring system will consume and configure their exporters accordingly; do not assume that choosing one exporter automatically enables every signal.
How do I check whether telemetry arrived?
- Start Claude Code with telemetry enabled and configured for the intended backend.
- In that backend, look for the
claude_code.session.countmetric after a session starts. - If you configured logs only, submit a prompt and look for the
claude_code.user_promptevent. - If neither appears, start Claude Code with
claude --debug-file <path>and inspect the resulting debug log for configuration or export errors.
Check the destination itself: a correctly set environment variable does not prove that data reached the collector or backend.
Rank #2
- FAST RUNS IN THE FAMILY — The 16-inch MacBook Pro with the M5 Pro or M5 Max chip brings next-generation speed and powerful on-device AI to personal, professional, and creative tasks. With all-day battery life, double the starting storage,* and a breathtaking Liquid Retina XDR display, it’s pro in every way.*
- BUCKLE UP — Along with a next-generation CPU, faster unified memory, and up to 2x faster SSD storage,* M5 Pro and M5 Max feature a more powerful GPU with a Neural Accelerator built into each core, delivering faster AI performance and on-device training capabilities. So you can blaze through demanding workloads at mind-bending speeds.
- BUILT FOR AI — Apple silicon, and every major component that powers it, is designed to run demanding on-device AI workloads like LLM inference and training. And Apple Intelligence helps you write, express yourself, and get things done effortlessly with groundbreaking privacy protections at every step.*
- ALL-DAY BATTERY LIFE — MacBook Pro delivers the same exceptional performance whether it’s running on battery or plugged in.*
- MACOS RUNS APPS FAST — All your go-to apps run lightning fast in macOS, including built-in apps like FaceTime and Messages. Plus, built-in virus protection and free software updates help keep your Mac running smoothly and securely.
How do I track usage across a team and attribute costs?
Administrators can distribute OpenTelemetry configuration through managed settings. To distinguish teams or cost centers in dashboards, set custom OTEL_RESOURCE_ATTRIBUTES values, using the documented comma-separated key=value format. For example, labels such as a department or team can support grouping, provided the values follow that strict formatting.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →In the documented setup, custom resource attributes become metric labels. Keep values low-cardinality: labels that vary for many individual users or sessions can multiply time series and increase storage costs in the monitoring backend. Choose labels that support meaningful organizational reporting without creating a distinct series for every changing value.
Telemetry detail is a privacy decision
Anthropic’s documented settings leave prompt and assistant-response content disabled by default. Other controls govern tool details and tool content; enabling raw API body events includes the entire conversation history. Before enabling richer content, decide what the organization needs to observe, who can access it, and which retention and access controls apply in the selected backend. Anthropic documents configuration options, not one retention policy that applies to every backend.
When is an LLM gateway useful?
A gateway is a separate architecture option when an organization needs a central place for usage attribution, budgets or rate limits, and audit logs. It can complement telemetry, but it is not the same as turning on OpenTelemetry. Anthropic cautions that “the gateway becomes infrastructure your organization operates”: the organization must run it and keep it compatible as Claude Code evolves. Anthropic also states that third-party gateways are not endorsed, maintained, or audited by it. See Anthropic’s gateway documentation before choosing that approach.
How do I add Claude Code to GitHub Actions?
Use Anthropic’s GitHub Action when you want Claude to respond to an @claude mention on an issue or pull request, or to run from a prompt supplied by a GitHub event. The documented manual setup is to install the Claude GitHub App, provide an API key or OAuth token as a GitHub secret, and add a workflow file under .github/workflows/. Teams can scale setup with organization-level secrets and reusable workflows. Follow the official GitHub Actions instructions for the current workflow syntax.
Choose credentials for the deployment
- API key: A practical choice for a shared secret across repositories; store it as a GitHub secret rather than embedding it in workflow text.
- Subscription OAuth token: Available for documented setups, but tied to the subscription of the person who generated it. Do not treat an individual’s token as an organization-wide credential.
- Workload identity federation: Use the documented configuration when the deployment is designed to authenticate this way.
The appropriate option depends on how the organization manages credentials and its deployment. Keep credentials out of commands or output that may be printed in workflow logs.
Use the CLI for a custom automation job
For scripts or a custom CI job that does not need the GitHub Action, Claude Code offers non-interactive mode through claude -p "query". This runs a prompt-driven invocation instead of opening the regular interactive session. The CLI reference documents controls including --allowedTools, --disallowedTools, JSON output formats and turn limits. Restrict tools to what the job needs and choose an output format that downstream steps can consume.
The action and custom CLI approach solve different workflow needs: the action provides GitHub-specific mention or event integration, while claude -p is a building block for scripts and other automation. Neither configures usage telemetry automatically.
Why doesn’t CI run after Claude pushes a commit?
Check the workflow trigger and the credential used to create the commit. Anthropic notes that GitHub does not trigger workflows on commits created using the default GITHUB_TOKEN. If later checks need to run on Claude-authored commits, confirm the CI workflow listens for the relevant events, such as push or pull_request, and review how the Claude workflow authenticates. The Claude action uses GitHub App authentication by default when github_token is omitted; a custom token can change the behavior. Avoid assuming a commit will trigger a second workflow without checking its event and authentication setup.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Why is the @claude workflow not responding?
Work through the GitHub-side prerequisites and event conditions:
- Confirm the Claude GitHub App is installed for the repository or organization in question.
- Check that GitHub Actions is enabled and the workflow file is present under
.github/workflows/. - Verify that the required secret is available to that workflow and that its credential is valid.
- Check that the issue or pull request comment contains the exact
@claudetrigger phrase expected by the workflow. - Confirm the triggering actor has write access where required. Anthropic documents write-access checks for issue and pull-request events, and bot actors are rejected unless explicitly allowed.
- Review the workflow’s trigger events and permissions to ensure the event is eligible to start it and that the job has only the repository access it needs.
These checks concern GitHub workflow execution; they do not diagnose missing telemetry. For telemetry export failures, use the backend verification steps and Claude debug file instead.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
