# AgentMark Docs

> AgentMark helps teams build reliable AI agents. Manage prompts, trace executions, run evaluations, and deploy agents with confidence.

## Docs

- [Create alert](https://docs.agentmark.co/api-reference/alerts/create-alert.md): Creates a threshold alert.
- [Delete alert](https://docs.agentmark.co/api-reference/alerts/delete-alert.md): Deletes an alert. Cascades to its alert_history rows.
- [Get alert](https://docs.agentmark.co/api-reference/alerts/get-alert.md): Returns a single alert by ID.
- [List alert trigger history](https://docs.agentmark.co/api-reference/alerts/list-alert-trigger-history.md): Returns the chronological trigger/resolve history for an alert, newest first. Each record captures the value that crossed the threshold and the commit SHA at the time, so agents can correlate alerts with deploys.
- [List alerts](https://docs.agentmark.co/api-reference/alerts/list-alerts.md): Returns alerts for the authenticated application, newest first.
- [List Slack channels available for alert notifications](https://docs.agentmark.co/api-reference/alerts/list-slack-channels-available-for-alert-notifications.md): Returns every channel the Slack bot installed for this app is a member of, in the order Slack returns them. Use this to discover a `channel_id` to wire into an alert with `use_slack: true`.
- [Update alert](https://docs.agentmark.co/api-reference/alerts/update-alert.md): Replaces an alert with the supplied body. The full alert state must be sent (PUT semantics) so the metric/evaluation field-coupling rule can be enforced in one validation pass. To toggle a single field, GET the alert, mutate, and PUT.
- [Add items to queue](https://docs.agentmark.co/api-reference/annotation-queue-items/add-items-to-queue.md): Adds one or more traces/spans/sessions to the queue for review. Duplicate `(queue_id, resource_id)` pairs are ignored silently.
- [Delete queue item](https://docs.agentmark.co/api-reference/annotation-queue-items/delete-queue-item.md): Removes an item from a queue. Cascades to reviewer records.
- [Get queue item](https://docs.agentmark.co/api-reference/annotation-queue-items/get-queue-item.md): Returns a single queue item by ID.
- [List queue items](https://docs.agentmark.co/api-reference/annotation-queue-items/list-queue-items.md): Returns every item enqueued for review, in the order they were added.
- [Submit review](https://docs.agentmark.co/api-reference/annotation-queue-items/submit-review.md): Submit a review (`completed` or `skipped`) on behalf of the authenticated user. When the queue's `reviewers_required` threshold is met, the item auto-advances to `completed`.
- [Update queue item](https://docs.agentmark.co/api-reference/annotation-queue-items/update-queue-item.md): Updates item status or assigned reviewer. Setting status to `completed` auto-records `completed_by` / `completed_at`.
- [Create annotation queue](https://docs.agentmark.co/api-reference/annotation-queues/create-annotation-queue.md): Creates a new queue for collecting human review on traces, spans, or sessions.
- [Delete annotation queue](https://docs.agentmark.co/api-reference/annotation-queues/delete-annotation-queue.md): Deletes a queue. Cascades to its items and reviewer records.
- [Get annotation queue](https://docs.agentmark.co/api-reference/annotation-queues/get-annotation-queue.md): Returns metadata for a single queue by ID.
- [List annotation queues](https://docs.agentmark.co/api-reference/annotation-queues/list-annotation-queues.md): Returns every annotation queue for the authenticated application, with per-queue progress counters (pending / in_progress / completed / skipped / total).
- [Update annotation queue](https://docs.agentmark.co/api-reference/annotation-queues/update-annotation-queue.md): Updates mutable queue metadata (`name`, `description`, `status`, `instructions`, `reviewers_required`, `score_config_names`). Fields not provided are left unchanged.
- [Create API key](https://docs.agentmark.co/api-reference/api-keys/create-api-key.md): Creates a new API key. The plaintext key is returned EXACTLY ONCE in `data.plaintext_key` — record it immediately. Subsequent reads expose only metadata.
- [List API keys](https://docs.agentmark.co/api-reference/api-keys/list-api-keys.md): Returns API keys for the authenticated tenant. Plaintext is never returned on this endpoint — record it at creation time.
- [Revoke API key](https://docs.agentmark.co/api-reference/api-keys/revoke-api-key.md): Revokes the API key and removes its local metadata row. Note: a brief revocation lag may occur — the gateway caches verified credentials for a short TTL, so a freshly-revoked key may still verify until the cache expires.
- [Clear the linked repository and branch](https://docs.agentmark.co/api-reference/apps/clear-the-linked-repository-and-branch.md): Clears `git_connection.repository` and removes the `git_branch` row. The OAuth install is preserved so the user can re-link without re-clicking the install URL.
- [Create app](https://docs.agentmark.co/api-reference/apps/create-app.md): Creates a new app in the authenticated tenant. `runtime` defaults to `nodejs` if omitted.
- [Delete app](https://docs.agentmark.co/api-reference/apps/delete-app.md): Deletes an app. Cascades to all child rows via ON DELETE CASCADE (git_connection, git_branch, app_connection, llm_api_url, alerts, deployments, etc.).
- [Get app](https://docs.agentmark.co/api-reference/apps/get-app.md): Returns a single app by ID.
- [Get git connection status for an app](https://docs.agentmark.co/api-reference/apps/get-git-connection-status-for-an-app.md): Returns the current git connection state. `connected: true` means a `git_connection` row exists (OAuth handshake completed). `repository` is null between OAuth-done and repo-picked.
- [Link a repository and branch to an app](https://docs.agentmark.co/api-reference/apps/link-a-repository-and-branch-to-an-app.md): Persists the chosen repository + branch onto the app's existing git_connection. The next push to `branch` triggers the deploy webhook which materializes templates and datasets.
- [List apps](https://docs.agentmark.co/api-reference/apps/list-apps.md): Returns apps for the authenticated tenant, newest first. Use `?name=<exact>` to look up a specific app by name without paginating.
- [List branches in a repository](https://docs.agentmark.co/api-reference/apps/list-branches-in-a-repository.md): Returns the list of branch names for `repository` (passed as a query param because the value contains a slash). Use after `list-app-git-repositories` to populate a branch picker before linking.
- [List repositories accessible to the app's git installation](https://docs.agentmark.co/api-reference/apps/list-repositories-accessible-to-the-apps-git-installation.md): Returns repositories the linked git installation can see. For GitHub, this is the App's `/installation/repositories`. For GitLab, the OAuth user's `membership=true` projects.
- [Mint an OAuth authorization URL for git-provider connect](https://docs.agentmark.co/api-reference/apps/mint-an-oauth-authorization-url-for-git-provider-connect.md): Returns a per-provider OAuth authorization URL plus a signed state token. Headless flow:
- [Update app](https://docs.agentmark.co/api-reference/apps/update-app.md): Updates writable fields on an app. PATCH semantics — any subset of `name`, `runtime`, `entry_point` may be sent. An empty body is rejected with 400.
- [Authentication](https://docs.agentmark.co/api-reference/authentication.md): How to authenticate with the AgentMark Gateway API.
- [Get capabilities](https://docs.agentmark.co/api-reference/capabilities/get-capabilities.md): Returns a map of available API endpoints for the current target (cloud or local). Use this to discover which features are supported before calling other endpoints.
- [Get config](https://docs.agentmark.co/api-reference/config/get-config.md): Returns the effective project configuration synced from `agentmark.json` for the authenticated application, plus the current synced commit SHA when available.
- [Append dataset row](https://docs.agentmark.co/api-reference/datasets/append-dataset-row.md): Appends a single row to the specified dataset. The `datasetName` parameter is the dataset path without the `.jsonl` extension, URL-encoded.
- [Import dataset rows from spans](https://docs.agentmark.co/api-reference/datasets/import-dataset-rows-from-spans.md): Transforms one or more spans into canonical dataset rows and appends them to the specified dataset.
- [Import dataset rows from traces](https://docs.agentmark.co/api-reference/datasets/import-dataset-rows-from-traces.md): Transforms one or more traces into canonical dataset rows and appends them to the specified dataset.
- [List datasets](https://docs.agentmark.co/api-reference/datasets/list-datasets.md): Returns a paginated list of datasets for your application with per-dataset metadata (`row_count`, `created_at`).
- [Get deployment](https://docs.agentmark.co/api-reference/deployments/get-deployment.md): Returns a single deployment by id. The response includes status (`deployment_status`, `files_status`, `code_status`), commit metadata, timing, and any failure reason — sufficient to monitor progress without needing the dedicated logs endpoint for most use cases. For deployments created by an environ…
- [List deployments](https://docs.agentmark.co/api-reference/deployments/list-deployments.md): Returns a paginated list of deployments for the authenticated application, newest first. Supports `?status=running|success|failure` to filter by deployment status — the typical CI use case ("is my deploy still going?") is `?status=running`.
- [Create environment](https://docs.agentmark.co/api-reference/environments/create-environment.md): Creates a new environment in the current app. The env starts in the no-pin state — promote (POST `/v1/environments/{id}/promote`) is the only path to pin a version. Synchronously provisions a per-env Fly app; failure rolls back the env row and surfaces 500 `env_fly_provisioning_failed`. Per FR-028 n…
- [Delete environment](https://docs.agentmark.co/api-reference/environments/delete-environment.md): Hard-deletes a non-default environment after a typed-name confirmation. Cascade order: env row → CASCADE removes api_key / alert / deployment / template_snapshot / env_config_snapshot / annotation_queue; SET NULL preserves alert_history (denormalized name + epoch kept intact). Fly app destroy is att…
- [Get environment](https://docs.agentmark.co/api-reference/environments/get-environment.md): Returns a single environment by id. Includes `cascade_preview` (counts of api_key / alert / deployment / template_snapshot rows that would be deleted on env delete — see FR-090) and `in_flight_saga_id` (id of the most-recent pending/snapshotting/deploying saga deployment row, or null).
- [List environments](https://docs.agentmark.co/api-reference/environments/list-environments.md): Returns environments for the authenticated application, default env first. Each env carries pin state (`current_version` — 0 when no-pin — plus `current_commit_sha`, the git commit of the pinned deployment, NULL when no-pin) and a stable `epoch` that distinguishes env instances across name reuse aft…
- [List the saga deployment audit log for an environment](https://docs.agentmark.co/api-reference/environments/list-the-saga-deployment-audit-log-for-an-environment.md): Returns every saga `deployment` row (`env_version IS NOT NULL`) for this env in reverse chronological order. Per FR-030/FR-043. Replaces the removed 054-era `GET /v1/environments/{id}/promotion-history`.
- [Promote to environment](https://docs.agentmark.co/api-reference/environments/promote-to-environment.md): Promotes the source environment's current content + code into this environment. Runs the unified deployment in-process: it writes the env-keyed `template_snapshot`, atomically commits the env pointer, AND dispatches the managed code build — this CLOSES the env-pin → code-deploy gap (a promote now de…
- [Roll back an environment](https://docs.agentmark.co/api-reference/environments/roll-back-an-environment.md): Rolls this environment back to a prior pinned version (FR-018..FR-023). `target_version` must name a successful forward-promote deployment row on this env — rollback re-runs the deployment at that row's `commit_sha`. Rollback is a promote at an older commit: it writes a fresh env-keyed snapshot, com…
- [Get baseline scores](https://docs.agentmark.co/api-reference/experiments/get-baseline-scores.md): Return per-(row × scorer) scores from the baseline run matching `commit_sha` (a content-addressed git tree hash), optionally narrowed by `dataset_path`. Used by `agentmark run-experiment --baseline-commit` to drive the regression gate: each score is keyed by `inputHash` (a stable hash of the row’s d…
- [Get experiment](https://docs.agentmark.co/api-reference/experiments/get-experiment.md): Retrieve a specific experiment by ID, including its per-item details (trace IDs, inputs/outputs, per-item cost/latency/tokens, and any scores attached to each trace).
- [List experiments](https://docs.agentmark.co/api-reference/experiments/list-experiments.md): Retrieve a paginated list of experiments (dataset runs). Each experiment is a group of traces that share a `DatasetRunId`, typically produced when a prompt is evaluated against every row in a dataset.
- [Files health](https://docs.agentmark.co/api-reference/health/files-health.md): Check the health of the files service and its dependencies.
- [Ingestion health](https://docs.agentmark.co/api-reference/health/ingestion-health.md): Check the health of the trace ingestion pipeline and its dependencies.
- [Service health](https://docs.agentmark.co/api-reference/health/service-health.md): Check if the gateway service is running. Returns healthy if all required environment variables are configured.
- [Get metrics](https://docs.agentmark.co/api-reference/metrics/get-metrics.md): Retrieve aggregated analytics metrics for your application. Returns a summary and an hourly time series for trace volume, latency, cost, token usage, and error rates.
- [API reference](https://docs.agentmark.co/api-reference/overview.md): Programmatic access to the AgentMark Gateway API.
- [Get LLM pricing](https://docs.agentmark.co/api-reference/pricing/get-llm-pricing.md): Returns per-model pricing for cost calculation. Response is a dynamic map keyed by model ID (e.g. `gpt-5`, `claude-opus-4-6`). Prices are per 1,000 tokens.
- [List or look up prompts](https://docs.agentmark.co/api-reference/prompts/list-or-look-up-prompts.md): List prompt files. With `?name=X`, filters to prompts whose frontmatter `name` matches — used by the trace drawer to map a span's `prompt_name` back to a file path. Without `name`, the full listing is OSS-only (cloud returns 501). Multiple matches are possible because frontmatter `name` is unique on…
- [Get score config](https://docs.agentmark.co/api-reference/score-configs/get-score-config.md): Returns a single score config by name. The name is the object key in `agentmark.json`'s `scores` map. Returns 404 if no config with that name is declared.
- [List score configs](https://docs.agentmark.co/api-reference/score-configs/list-score-configs.md): Returns the score configs declared in the application's `agentmark.json` (synced on deploy). Read-only — to add or modify a config, edit `agentmark.json` and redeploy. Returns an empty list if no configs are declared.
- [Create score](https://docs.agentmark.co/api-reference/scoring/create-score.md): Create a score record for a span or trace. Scores are used to track quality metrics, evaluation results, and human feedback.
- [Create scores (batch)](https://docs.agentmark.co/api-reference/scoring/create-scores-batch.md): Create up to 1000 scores in a single request. Each item is validated independently and the response always contains a per-item results array.
- [Delete score](https://docs.agentmark.co/api-reference/scoring/delete-score.md): Delete a score record by ID.
- [Get score](https://docs.agentmark.co/api-reference/scoring/get-score.md): Retrieve a single score record by ID. Returns the full score object including its value, label, reason, and source.
- [Get score aggregations](https://docs.agentmark.co/api-reference/scoring/get-score-aggregations.md): Returns aggregated statistics for scores grouped by name. Useful for understanding score distributions across your application.
- [Get score names](https://docs.agentmark.co/api-reference/scoring/get-score-names.md): Returns a list of distinct score names used in your application. Useful for building filter dropdowns and discovering available score types.
- [List scores](https://docs.agentmark.co/api-reference/scoring/list-scores.md): Returns a paginated list of scores for the authenticated application. Supports filtering by resource, name, source, and date range.
- [Search scores](https://docs.agentmark.co/api-reference/scoring/search-scores.md): Search scores with structured JSON filters. `filters` is an AND-list of predicates (`{field, operator, value}`) and one-level OR-groups (`{or: [...]}`) over the score fields: name, score, source, user_id, resource_id, label, created_at. Membership (`in`, `notIn`) and range (`between`) operators are…
- [Get filter schema](https://docs.agentmark.co/api-reference/search/get-filter-schema.md): Machine-readable description of the filterable surface for `POST /v1/traces/search`, `POST /v1/spans/search`, `POST /v1/scores/search`, and the `filter` string DSL on the GET list endpoints: fields and operators per resource, dynamic field patterns (`metadata.<key>`, `score__<name>`), and per-reques…
- [List sessions](https://docs.agentmark.co/api-reference/sessions/list-sessions.md): Retrieve a paginated list of sessions. Sessions group related traces together for multi-turn conversations, workflows, and batch processing.
- [Get span](https://docs.agentmark.co/api-reference/spans/get-span.md): Fetch a single span by its globally-unique span_id, without needing its trace_id. Returns the full span: metadata plus its input/output payload (the span analog of GET /v1/traces/:traceId).
- [Get span I/O detail](https://docs.agentmark.co/api-reference/spans/get-span-io-detail.md): Returns the full input/output payload for a specific span, plus parsed output objects, tool calls, and the span’s custom metadata (reserved internal namespaces excluded). Useful for rendering a single span in isolation without loading the entire trace.
- [List spans](https://docs.agentmark.co/api-reference/spans/list-spans.md): Query spans across all traces. Supports filtering by type, status, model, name, and duration range.
- [List spans for a trace](https://docs.agentmark.co/api-reference/spans/list-spans-for-a-trace.md): Returns every span belonging to the given trace, ordered by start time. Not paginated — traces are bounded by span volume (capped at ingest), so the full list is returned in a single response.
- [Search spans](https://docs.agentmark.co/api-reference/spans/search-spans.md): Search spans with structured JSON filters — the programmatic form of the `filter` string DSL on `GET /v1/spans`. `filters` is an AND-list of predicates (`{field, operator, value}`) and one-level OR-groups (`{or: [...]}`). Adds membership operators `in`, `notIn`, and `between`. Valid fields and opera…
- [Get template](https://docs.agentmark.co/api-reference/templates/get-template.md): Retrieve a prompt template by its file path. Templates must have a `.mdx` or `.jsonl` extension.
- [Get trace](https://docs.agentmark.co/api-reference/traces/get-trace.md): Retrieve a specific trace by ID, including all its spans.
- [Ingest traces](https://docs.agentmark.co/api-reference/traces/ingest-traces.md): Ingest trace data in [OTLP (OpenTelemetry Protocol)](https://opentelemetry.io/docs/specs/otlp/) format. Traces are buffered in a queue and processed asynchronously.
- [List traces](https://docs.agentmark.co/api-reference/traces/list-traces.md): Retrieve a paginated list of traces. Supports filtering by status, user, model, and date range.
- [Search traces](https://docs.agentmark.co/api-reference/traces/search-traces.md): Search traces with structured JSON filters — the programmatic form of the `filter` string DSL on `GET /v1/traces`. `filters` is an AND-list of predicates (`{field, operator, value}`) and one-level OR-groups (`{or: [...]}`). Adds membership operators `in`, `notIn`, and `between` that have no string-D…
- [API versioning & stability](https://docs.agentmark.co/api-reference/versioning.md): How the AgentMark Gateway API versions itself, what counts as a breaking change, and how deprecations are communicated.
- [Components](https://docs.agentmark.co/build/components.md): Create reusable components to share prompting patterns across your prompts
- [Create a prompt](https://docs.agentmark.co/build/creating-prompts.md): Create prompts in the Dashboard visual editor or as local .prompt.mdx files
- [Example prompts](https://docs.agentmark.co/build/example-prompts.md): Copy-paste starter prompts that cover all four AgentMark generation types (object, text + tools, image, and speech), plus a runnable dataset for each.
- [File attachments](https://docs.agentmark.co/build/file-attachments.md): Attach images and files to prompts using the ImageAttachment and FileAttachment tags.
- [Generating images](https://docs.agentmark.co/build/generation-types/image.md): Generate images from prompts using AgentMark with DALL-E, Stable Diffusion, and other image models.
- [Generating objects](https://docs.agentmark.co/build/generation-types/object.md): Generate structured JSON objects from prompts with schema validation using AgentMark.
- [Generation types overview](https://docs.agentmark.co/build/generation-types/overview.md): Understand the different types of content you can generate with AgentMark prompts
- [Generating speech](https://docs.agentmark.co/build/generation-types/speech.md): Generate speech audio from prompts using AgentMark with text-to-speech models.
- [Generating text](https://docs.agentmark.co/build/generation-types/text.md): Generate text completions from prompts using AgentMark with any LLM provider.
- [MCP integration](https://docs.agentmark.co/build/mcp.md): Reference Model Context Protocol (MCP) tools by name in AgentMark prompts, then connect the servers at your own call site.
- [Build](https://docs.agentmark.co/build/overview.md): Create, run, and version prompts in code or in the Dashboard
- [Playground](https://docs.agentmark.co/build/playground.md): Compare prompts and models side-by-side to find the best configuration before publishing
- [Running prompts](https://docs.agentmark.co/build/running-prompts.md): Run prompts from the Dashboard, CLI, or SDK with streaming, tracing, and caching
- [Schema references](https://docs.agentmark.co/build/schema-references.md): Reuse JSON schema definitions across AgentMark prompts with $ref
- [Prompt syntax](https://docs.agentmark.co/build/syntax.md): Learn the AgentMark template syntax for creating dynamic prompts
- [Tools and agents](https://docs.agentmark.co/build/tools-and-agents.md): Reference tools by name in prompts, resolve them to implementations at your call site, and build multi-step agent workflows
- [Version control](https://docs.agentmark.co/build/version-control.md): Track, compare, and manage prompt versions in the AgentMark Dashboard
- [Client config](https://docs.agentmark.co/configure/client-config.md): Configure your AgentMark client for prompt loading, tools, evals, and tracing with one neutral pattern that works with any SDK.
- [Environment variables](https://docs.agentmark.co/configure/environment-variables.md): Complete reference for all environment variables used by AgentMark
- [Loaders](https://docs.agentmark.co/configure/loaders.md): Load prompts from different sources using ApiLoader and FileLoader
- [Model schemas](https://docs.agentmark.co/configure/model-schemas.md): Add built-in and custom models in AgentMark
- [Project config](https://docs.agentmark.co/configure/project-config.md): How to configure AgentMark for your application
- [Type safety](https://docs.agentmark.co/configure/type-safety.md): Generate TypeScript types from your prompt schemas for compile-time validation and autocomplete.
- [API keys](https://docs.agentmark.co/deploy/api-keys.md): Create, scope, and manage API keys for the AgentMark Gateway from the Dashboard.
- [Billing and usage](https://docs.agentmark.co/deploy/billing-and-usage.md): Pricing tiers, usage limits, entitlements, and rate limits
- [Deployment](https://docs.agentmark.co/deploy/deployment.md): Connect a Git repository and AgentMark Cloud ships your prompts and handler code on every push.
- [Environments and promotions](https://docs.agentmark.co/deploy/environments-and-promotions.md): Run the same app at different versions in isolated environments, and promote a tested version forward when it's ready.
- [Organizations and teams](https://docs.agentmark.co/deploy/organizations-and-teams.md): Organizations, apps, and how they fit together
- [Regression gates](https://docs.agentmark.co/deploy/regression-gates.md): Fail a PR or build when experiment scores regress against a baseline run
- [Security](https://docs.agentmark.co/deploy/security.md): How AgentMark protects your data: SSO, encryption, PII masking, and data residency
- [Status checks](https://docs.agentmark.co/deploy/status-checks.md): Fail pull/merge requests when AgentMark prompts don't compile, the same way you'd fail them for a broken build.
- [Users and access control](https://docs.agentmark.co/deploy/users-and-access-control.md): Roles, permissions, custom roles, app-level access, and team management
- [Webhooks](https://docs.agentmark.co/deploy/webhooks.md): Receive alert notifications from AgentMark Cloud via webhooks
- [Human annotation](https://docs.agentmark.co/evaluate/annotations.md): Score traces with human reviewers, individually or in structured batch queues
- [Datasets](https://docs.agentmark.co/evaluate/datasets.md): Create test datasets for your prompts
- [GitLab CI/CD](https://docs.agentmark.co/evaluate/gitlab-ci.md): Run AgentMark evals on changed prompts and gate merge requests on the results
- [Evaluate](https://docs.agentmark.co/evaluate/overview.md): Test and improve your prompts with datasets, evaluators, experiments, and annotations
- [Running experiments](https://docs.agentmark.co/evaluate/running-experiments.md): Test prompts with datasets in the Dashboard, CLI, or SDK
- [Evaluations](https://docs.agentmark.co/evaluate/writing-evals.md): Write evaluation functions to score prompt outputs
- [AgentMark agent skill](https://docs.agentmark.co/getting-started/agent-skill.md): Install the AgentMark skill so Claude Code, Codex, Cursor, or GitHub Copilot already know how to author prompts, run experiments, and deploy in any project, including ones you did not scaffold with create agentmark
- [Set up your AgentMark client](https://docs.agentmark.co/getting-started/client-setup.md): Create your AgentMark client, run prompts and experiments locally, and deploy it so the Dashboard can execute against your code.
- [Quickstart](https://docs.agentmark.co/getting-started/quickstart.md): Create an AgentMark project and run your first prompt. Your AI editor does the wiring.
- [Bring your own SDK](https://docs.agentmark.co/integrations/bring-your-own-sdk.md): Migrate an app built on any LLM SDK (raw AWS Bedrock, OpenAI, a bespoke client) onto AgentMark for prompt management, tracing, and experiments, without rewriting your model calls.
- [Overview](https://docs.agentmark.co/integrations/overview.md): Connect AgentMark prompts to any LLM SDK with one pattern, a neutral prompt render plus an optional executor. Reference executors for Vercel AI SDK, Pydantic AI, Mastra, and the Claude Agent SDK.
- [Reference executors](https://docs.agentmark.co/integrations/reference-executors.md): Complete, self-contained executor implementations for the Vercel AI SDK, the raw OpenAI and Anthropic clients, and agent frameworks. Copy one into your project; no AgentMark adapter package required.
- [Core concepts](https://docs.agentmark.co/introduction/core-concepts.md): AgentMark's core building blocks: organizations, apps, branches, environments, prompts, traces, datasets, evals, sessions, alerts, and annotations, and how they fit together.
- [What is AgentMark?](https://docs.agentmark.co/introduction/overview.md): AgentMark helps teams build reliable AI agents. Manage prompts, trace executions, run evaluations, and deploy, locally or with AgentMark Cloud.
- [Alerts](https://docs.agentmark.co/observe/alerts.md): Monitor your application with customizable alerts
- [Cost and token tracking](https://docs.agentmark.co/observe/cost-and-token-tracking.md): Monitor LLM spending and token usage across your application
- [Dashboards](https://docs.agentmark.co/observe/dashboards.md): Monitor operational metrics, evaluation scores, and custom analytics in one place
- [Filtering and search](https://docs.agentmark.co/observe/filtering-and-search.md): Find specific traces and sessions using filters, date ranges, sorting, saved views, and shareable URLs
- [Metadata](https://docs.agentmark.co/observe/metadata.md): Attach custom key-value pairs to traces for filtering, debugging, and context
- [Observe](https://docs.agentmark.co/observe/overview.md): Monitor, debug, and optimize your LLM applications with tracing, dashboards, and alerts
- [PII masking](https://docs.agentmark.co/observe/pii-masking.md): Redact sensitive data from traces before they leave your application
- [Sessions](https://docs.agentmark.co/observe/sessions.md): Group related traces together for multi-turn conversations, workflows, and batch processing
- [Tags](https://docs.agentmark.co/observe/tags.md): Attach string labels to traces for categorization and filtering
- [Traces and logs](https://docs.agentmark.co/observe/traces-and-logs.md): Monitor and debug your prompts with distributed tracing
- [Tracing setup](https://docs.agentmark.co/observe/tracing-setup.md): Instrument your application with the AgentMark SDK to capture traces, spans, and metrics
- [Troubleshooting traces](https://docs.agentmark.co/observe/troubleshooting.md): Why your AgentMark traces aren't showing up, the silent failures that drop spans, and how to fix each one.
- [CLI reference](https://docs.agentmark.co/reference/cli-commands.md): Complete reference for all AgentMark CLI commands
- [MCP servers](https://docs.agentmark.co/reference/mcp-servers.md): Wire AgentMark MCP servers into your AI editor: the docs MCP for authoring .prompt.mdx files, and the gateway MCP for driving the full AgentMark API
- [Python dev server](https://docs.agentmark.co/reference/python-dev-server.md): Running the AgentMark development server with Python
- [Troubleshooting](https://docs.agentmark.co/reference/troubleshooting.md): Solutions to common errors and issues when using AgentMark
- [Components](https://docs.agentmark.co/templatedx/components.md): Create reusable template fragments in TemplateDX for shared prompt logic and structure.
- [Expressions](https://docs.agentmark.co/templatedx/expressions.md): Use JavaScript-like expressions in TemplateDX templates for dynamic content and conditional logic.
- [Filters](https://docs.agentmark.co/templatedx/filters.md): Transform and format template values with built-in TemplateDX filter functions.
- [Overview](https://docs.agentmark.co/templatedx/introduction.md): TemplateDX is an extensible templating engine for AI prompts, built on Markdown and JSX.
- [Quickstart](https://docs.agentmark.co/templatedx/quickstart.md): Install TemplateDX and render your first prompt template.
- [Syntax overview](https://docs.agentmark.co/templatedx/syntax.md): Complete syntax reference for TemplateDX
- [Editor integration](https://docs.agentmark.co/templatedx/syntax-highlighting.md): Set up syntax highlighting, schema validation, and MCP integration for TemplateDX files in VS Code, Cursor, Zed, and Claude Code.
- [Tags](https://docs.agentmark.co/templatedx/tags.md): Use built-in and custom tags in TemplateDX to control template flow and transform data.
- [Variables](https://docs.agentmark.co/templatedx/variables.md): Pass dynamic data into TemplateDX templates via the props object.

## OpenAPI Specs

- [dashboard-openapi](https://docs.agentmark.co/dashboard-openapi.yaml)
- [openapi](https://docs.agentmark.co/openapi.yaml)

## Optional

- [GitHub](https://github.com/agentmark-ai/agentmark)
- [LinkedIn](https://www.linkedin.com/company/agentmark)