# Anectico > Person-centric observability for humans and AI agents. One identity connects errors, traces, logs, replay, product events, and agent runs. Canonical documentation: https://anectico.com/docs/ Machine-readable manifest: https://anectico.com/docs/manifest.json Complete documentation corpus: https://anectico.com/llms-full.txt The links below return clean Markdown. Prefer the task page that matches the question; use reference pages for exact APIs, scopes, SDK calls, and limits. ## Start here - [Quickstart: see your first customer story](https://anectico.com/docs/start/quickstart.md): Connect one application, identify a test customer, and follow an error across their Anectico timeline. - [How Anectico connects the story](https://anectico.com/docs/start/how-anectico-works.md): Understand what the SDK sends, how customer identity links signals, and where to begin in the product. - [Choose an integration](https://anectico.com/docs/start/choose-integration.md): Pick the shortest supported path for your application and see which signals each SDK provides. - [Verify your setup](https://anectico.com/docs/start/verify-setup.md): Confirm that telemetry and customer identity are connected before expanding your rollout. ## Instrument - [Identify customers](https://anectico.com/docs/instrument/identity.md): Use one stable customer identity across errors, traces, logs, replay, events, and AI calls. - [JavaScript and TypeScript](https://anectico.com/docs/instrument/javascript.md): Instrument Node.js, browser, Express, Fastify, NestJS GraphQL, Prisma, and React applications with the Anectico SDK. - [Python](https://anectico.com/docs/instrument/python.md): Instrument Python services with metrics, telemetry, errors, and request-safe customer identity. - [Go](https://anectico.com/docs/instrument/go.md): Instrument Go services with metrics, traces, logs, errors, and request-scoped customer identity. - [Choose a mobile SDK](https://anectico.com/docs/instrument/mobile.md): Connect iOS, Android, React Native, or Flutter activity to the same customer story. - [iOS](https://anectico.com/docs/instrument/ios.md): Capture iOS identity, events, errors, crashes, and backend request context. - [Android](https://anectico.com/docs/instrument/android.md): Capture Android identity, events, errors, crashes, and backend request context. - [React Native](https://anectico.com/docs/instrument/react-native.md): Connect JavaScript and native mobile failures to one React Native customer journey. - [Flutter](https://anectico.com/docs/instrument/flutter.md): Connect Dart and native mobile failures to one Flutter customer journey. - [OpenTelemetry](https://anectico.com/docs/instrument/opentelemetry.md): Send existing OTLP traces, logs, and metrics to Anectico without replacing your instrumentation. - [Record application metrics](https://anectico.com/docs/instrument/metrics.md): Choose the right metric type, emit low-cardinality measurements, and verify that Anectico can query them. - [Capture errors and releases](https://anectico.com/docs/instrument/errors.md): Send actionable errors, keep grouping useful, and resolve production stack frames. - [Record session replay safely](https://anectico.com/docs/instrument/session-replay.md): Capture browser sessions with privacy masking and connect them to errors and backend traces. - [Record LLM calls](https://anectico.com/docs/instrument/llm-calls.md): Capture model, token, latency, error, and cost data and attach it to the customer who caused it. - [Capture product events](https://anectico.com/docs/instrument/diagnostic-events.md): Measure application milestones, declare sampling, and inspect capture quality. - [Identity correction and instrumentation doctor](https://anectico.com/docs/instrument/diagnostics.md): Inspect identity links, separate mistaken aliases, and diagnose trace and log visibility. ## Investigate - [Understand one customer’s story](https://anectico.com/docs/investigate/customer-story.md): Use the customer timeline to follow errors, traces, logs, replay, events, and AI activity in order. - [Triage an Issue](https://anectico.com/docs/investigate/triage-issue.md): Prioritize an error by customer impact, reproduce its story, and move it through its lifecycle. - [Investigate with Customer Detective](https://anectico.com/docs/investigate/customer-detective.md): Ask production questions and verify grounded answers against cited telemetry. - [Search logs, traces, and metrics](https://anectico.com/docs/investigate/search-telemetry.md): Use raw telemetry when a customer story or Issue requires a deeper service-level investigation. - [Explore and compare metrics](https://anectico.com/docs/investigate/metrics.md): Query a metric with the right aggregation, build reusable charts, and connect a change to explanatory evidence. - [Investigate service health and dashboards](https://anectico.com/docs/investigate/service-health-and-dashboards.md): Find a broad service symptom, follow it into traces and logs, and save only the views worth repeating. - [Investigate with session replay](https://anectico.com/docs/investigate/session-replay.md): Use a browser recording, timeline markers, console, network, and errors without losing the customer context. - [Inspect events and customer cohorts](https://anectico.com/docs/investigate/events-and-cohorts.md): Check who performed a diagnostic event and save reusable customer groups for investigation or rollout targeting. - [Investigate accounts and segments](https://anectico.com/docs/investigate/accounts-and-segments.md): Connect customers to companies or workspaces and watch important account groups. - [Investigate agent runs](https://anectico.com/docs/investigate/agent-runs.md): Triage AI executions by customer, status, cost, tools, and quality evidence. - [Review detected anomalies](https://anectico.com/docs/investigate/anomalies.md): Triage unusual metric behavior, confirm impact, and close findings without mistaking detection for cause. - [Explain LLM spend per customer](https://anectico.com/docs/investigate/llm-spend.md): Find which customers, models, and agent runs are driving cost and failures. - [Check rollout health](https://anectico.com/docs/investigate/rollout-health.md): Connect feature exposure and releases to the customers and failures that followed. - [Discover event property schemas](https://anectico.com/docs/investigate/event-schema.md): Inspect permitted property names, JSON types and quality counts through MCP or the CLI. - [Manage tracking plans](https://anectico.com/docs/investigate/tracking-plans.md): Validate, preview and manage versioned expected instrumentation through MCP, the CLI and the dashboard. - [Investigate customer problem reports](https://anectico.com/docs/investigate/problem-reports.md): Collect a customer's description and connect it to their diagnostic evidence. - [Compare affected and successful customers](https://anectico.com/docs/investigate/customer-comparison.md): Find recorded attributes that differ between customers with an Issue and successful controls. - [Investigate click friction and release impact](https://anectico.com/docs/investigate/customer-experience.md): Connect observed click friction and release outcomes to people, accounts, operations and replay. - [Measure product events](https://anectico.com/docs/investigate/product-analytics.md): Execute typed trends, person funnels and retention and inspect frozen results and exact contributors through MCP or the CLI. - [Manage saved insights](https://anectico.com/docs/investigate/saved-insights.md): Create, inspect, revise and retire reusable analytics definitions through MCP, the CLI and the dashboard. - [Watch a saved product metric](https://anectico.com/docs/investigate/metric-watches.md): Create and validate a metric watch through MCP, REST, CLI or the dashboard using a pinned saved-insight revision and current source permissions. ## Respond - [Create useful alerts](https://anectico.com/docs/respond/alerts.md): Notify the right people about customer-impacting failures without creating noise. - [Manage incidents and on-call](https://anectico.com/docs/respond/incidents-and-on-call.md): Turn an active customer-impacting problem into an owned, traceable response. ## Manage Anectico - [Manage an organization and projects](https://anectico.com/docs/manage/workspace-and-projects.md): Keep customer data separated by project and manage organization access safely. - [Connect tools and notification delivery](https://anectico.com/docs/manage/connections-and-notifications.md): Connect external tools, create delivery channels, and verify alerts can reach responders. - [Review usage and AI settings](https://anectico.com/docs/manage/usage-and-ai-settings.md): Understand plan consumption, retention, LLM pricing, and automatic agent-run checks. - [Save searches and export data](https://anectico.com/docs/manage/save-and-export-data.md): Preserve repeatable filters and create bounded CSV, JSON, or Parquet export jobs. - [Decide what recorded content may be used for](https://anectico.com/docs/manage/content-policy.md): The per-project content policy: six kinds of content against four boundaries, which cells are enforced today, and the one denial that quietly empties two other boundaries. ## AI agents - [Run an agent without the dashboard](https://anectico.com/docs/agents/zero-dashboard-quickstart.md): Take an AI agent from no credential to a confirmed change, entirely from a terminal and the agent itself. - [Connect an AI agent with MCP](https://anectico.com/docs/agents/connect-mcp.md): Give Claude Code, Cursor, Codex, or another MCP client governed access to Anectico evidence. - [Use the Anectico CLI](https://anectico.com/docs/agents/cli.md): Investigate and operate Anectico from a terminal or coding agent. - [Permissions and safe actions](https://anectico.com/docs/agents/permissions-and-safety.md): Scope agent credentials, review evidence, and require confirmation for destructive changes. - [Scope recipes for agent jobs](https://anectico.com/docs/agents/scope-recipes.md): The minimum scope set for each common agent job, what each scope buys, and what to add when a job needs more. - [How agent access is enforced](https://anectico.com/docs/agents/security-model.md): Understand scopes, project binding, OAuth limits, profiles, confirmation, and audit coverage. - [Revoke an agent's access](https://anectico.com/docs/agents/revoke-agent-access.md): Identify and revoke an API key, verify that access stopped, and review what the agent already did. - [Connect an agent with A2A](https://anectico.com/docs/agents/connect-a2a.md): Let another AI agent delegate an investigation to Anectico over the Agent2Agent protocol. - [Evaluate an agent run](https://anectico.com/docs/agents/quality-evaluations.md): Create a deterministic evaluator, score a recorded run, and verify the result before enabling automatic checks. - [See every agent that is running](https://anectico.com/docs/agents/agent-inventory.md): The fleet and the asset inventory: which agents somebody declared, which ones Anectico found on its own, what needs review, and which credential each one is using. - [Drive one registered agent](https://anectico.com/docs/agents/drive-an-agent.md): Send one explicit A2A message to an agent you registered and inspect the recorded outcome. - [Replay a real recorded session](https://anectico.com/docs/agents/replay-a-recorded-session.md): Send a real person's recorded user turns to a registered agent and compare its new responses with the originals. - [Memory provenance and quarantine](https://anectico.com/docs/agents/memory-provenance.md): Give every memory item a stable identity, record a reviewed trust label for it, and see everywhere it flowed before and after that decision. - [Record context assembly](https://anectico.com/docs/agents/context.md): Record metadata-only evidence for the sources, compactions, and cache outcomes behind an agent model call. - [Contain an agent](https://anectico.com/docs/agents/contain-an-agent.md): Request, approve, verify, and lift an agent quarantine without overlooking failed enforcement. ## Reference - [Configuration](https://anectico.com/docs/reference/configuration.md): Common SDK and OpenTelemetry settings, defaults, and production guidance. - [Authentication and API keys](https://anectico.com/docs/reference/authentication.md): Create, scope, send, rotate, and revoke Anectico credentials safely. - [Permission scopes](https://anectico.com/docs/reference/permissions.md): Choose the least API-key permissions required for ingestion, investigation, response, and administration. - [REST API](https://anectico.com/docs/reference/rest-api.md): Use Anectico's public API conventions and find every supported customer resource family. - [Metrics REST API](https://anectico.com/docs/reference/metrics-api.md): Use the supported metric discovery, aggregation, histogram, forecast, and deletion endpoints. - [JavaScript SDK API](https://anectico.com/docs/reference/javascript-sdk.md): Functions, classes, options, parameters, and return values for the Anectico JavaScript and TypeScript SDK. - [Python SDK API](https://anectico.com/docs/reference/python-sdk.md): Classes, methods, parameters, return values, configuration, and integrations for the Anectico Python SDK. - [Go SDK API](https://anectico.com/docs/reference/go-sdk.md): Interfaces, functions, options, parameters, return values, and helpers for the Anectico Go SDK. - [iOS SDK API](https://anectico.com/docs/reference/ios-sdk.md): Functions, options, parameters, return values, and propagation helpers for the Anectico Swift SDK. - [Android SDK API](https://anectico.com/docs/reference/android-sdk.md): Functions, options, parameters, return values, and propagation helpers for the Anectico Kotlin SDK. - [React Native SDK API](https://anectico.com/docs/reference/react-native-sdk.md): Functions, options, parameters, promises, and native behavior for the Anectico React Native SDK. - [Flutter SDK API](https://anectico.com/docs/reference/flutter-sdk.md): Functions, options, parameters, futures, and native behavior for the Anectico Flutter SDK. - [CLI command reference](https://anectico.com/docs/reference/cli.md): Look up every Anectico CLI command, argument, flag, output convention, and exit code. - [MCP tool reference](https://anectico.com/docs/reference/mcp-tools.md): Understand Anectico's MCP tool families and discover the tools available to your key. - [Plans, limits, and retention](https://anectico.com/docs/reference/limits.md): Understand Anectico's planned usage meters and history windows during early access. - [MCP client compatibility](https://anectico.com/docs/reference/mcp-compatibility.md): What the Anectico MCP endpoint requires of a client, which protocol revisions it serves, and which hosts have been checked. - [Verify a signed attestation offline](https://anectico.com/docs/reference/verify-attestations.md): Check an exported agent bill of materials or action-receipt checkpoint yourself, with no call to Anectico. - [Agent span and event attributes](https://anectico.com/docs/reference/agent-attributes.md): The identifier rules and Anectico-defined attributes for agent spans, context events, guardrail decisions, and A2A calls. - [Machine-readable documentation](https://anectico.com/docs/reference/machine-readable-docs.md): Choose the Markdown, manifest, or MCP documentation surface that fits an agent, crawler, or retrieval pipeline. - [Evaluator configuration](https://anectico.com/docs/reference/evaluators.md): Configure evaluator kinds, exact versions, subjects, sampling, and paid-evaluation budgets. - [Evaluation results and human judgments](https://anectico.com/docs/reference/evaluation-results.md): Read scores, provenance, human annotations, calibration, and release series. - [Evaluation datasets and sandbox replay](https://anectico.com/docs/reference/evaluation-datasets.md): Freeze reviewed cases into dataset versions and interpret sandbox replay results. - [Evaluation experiments](https://anectico.com/docs/reference/evaluation-experiments.md): Compare recorded configurations on the same cases and interpret paired results. - [Evaluation release gates](https://anectico.com/docs/reference/evaluation-release-gates.md): Declare release objectives, evaluate admissible evidence, and interpret permanent decisions. - [Evaluation API and permissions](https://anectico.com/docs/reference/evaluation-api.md): Look up evaluation endpoints, scopes, and the supported CLI and MCP surface. - [Credential revocation and audit behavior](https://anectico.com/docs/reference/credential-revocation.md): Check revocation timing, streaming cutoffs, export ownership, audit limits, and delegation history. ## Help - [Troubleshooting](https://anectico.com/docs/help/troubleshooting.md): Start with the symptom and verify project, environment, time, identity, and delivery in order. - [No data is appearing](https://anectico.com/docs/help/no-data.md): Diagnose missing traces, logs, metrics, errors, events, and replay without changing unrelated settings. - [Customer activity is not connecting](https://anectico.com/docs/help/identity-not-linking.md): Fix missing, split, or incorrectly attributed customer timelines. - [Replay or symbols are not working](https://anectico.com/docs/help/replay-and-symbols.md): Diagnose missing browser recordings and unreadable JavaScript, Android, or iOS stack frames. - [Support, security, and privacy](https://anectico.com/docs/help/support.md): Contact Anectico, report a security issue, and ask for current trust and data-handling information. - [Authentication or access is failing](https://anectico.com/docs/help/authentication-and-access.md): Resolve sign-in, API-key, project, and missing-scope failures without widening access unnecessarily. - [MCP or CLI is not working](https://anectico.com/docs/help/agents-and-cli.md): Diagnose agent connection, CLI authentication, project selection, and safe-write failures. - [Alerts or integrations are not delivering](https://anectico.com/docs/help/alerts-and-connections.md): Trace a missing notification from rule evaluation through channel, connection, and on-call routing.