> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentmark.co/llms.txt
> Use this file to discover all available pages before exploring further.

# Changelog

> New features and improvements to AgentMark

<Update label="2026-06-22" description="Generated Images & Audio Render in Traces">
  ### Features & Improvements

  * **Generated images and audio render inline in traces** — when a span's output is generated media, the trace's **Input / Output** tab now shows the image or an audio player directly, instead of a wall of raw base64. Media is detected from the `{ mimeType, base64 }` shape (the AI SDK's `mediaType` key is also recognized). See [Generated media](/observe/traces-and-logs#generated-media).
  * **Oversized span fields are no longer truncated** — large outputs, inputs, tool calls, and base64 media are stored in object storage at ingest and fetched on demand when you open a span, so a big generation is preserved in full (previously it was clipped to fit the ingestion limit) while the trace list and timeline stay fast.
  * **Tracing your own image generation** — the AI SDK's `experimental_generateImage` emits no telemetry of its own; the new [Tracing image generation](/build/generation-types/overview#tracing-image-generation) guide shows how to capture generated media on a span from your application code.
</Update>

<Update label="2026-06-04" description="Custom Metadata in the Read API + String Filter Grammar">
  ### Features & Improvements

  * **Custom metadata is now readable via the API** — `GET /v1/traces/{traceId}` returns a trace-level `metadata` object plus per-span `metadata`; `GET /v1/traces/{traceId}/spans/{spanId}` and `GET /v1/spans` return span `metadata` alongside the I/O. Whatever you attach at ingestion (OTLP `agentmark.metadata.*` attributes or the SDK `metadata` option) is now retrievable programmatically and through the [`agentmark-mcp`](/ai-editors/mcp-servers) trace tools. Reserved internal namespaces (for example, `graph.node.*`) are excluded from the public object.
  * **Filter by `metadata.*` over the API** — the `filter` parameter on `GET /v1/traces` and `GET /v1/spans` now honors `metadata.<key>` predicates with `exists`, `does not exist`, `=`, and the string operators. See the [filter grammar](/api-reference/overview).
  * **String filter grammar** — `filter` is now a readable expression, for example `filter=metadata.env = "prod" and status = ERROR`. A malformed or unsupported filter returns a `400` with a clear message instead of being silently ignored.

  ### Breaking Changes

  * The `filter` parameter no longer accepts the previous JSON-array form; use the [string grammar](/api-reference/overview).
</Update>

<Update label="2026-04-15" description="Local/Cloud API Parity, New API Endpoints for Datasets, Experiments, Prompts, and Runs">
  ### Features & Improvements

  * **Local/cloud API parity** — The local dev server now supports the same API endpoints as the cloud gateway, so you can develop and test API integrations without a cloud account
  * New API endpoints for [datasets](/api-reference/overview), [experiments](/api-reference/overview), [prompt logs](/api-reference/overview), and [runs](/api-reference/overview) — query evaluation data programmatically via REST API or `agentmark api` CLI
  * New [`capabilities`](/api-reference/overview) endpoint to check which features a server supports, useful when writing code that targets both local and cloud environments
  * The `agentmark api` CLI command now supports `datasets`, `experiments`, `prompts`, `runs`, and `capabilities` resources
</Update>

<Update label="2026-04-11" description="REST API Reference, Score Analytics Widgets, PII Masking">
  ### Features & Improvements

  * New [API Reference](/api-reference/overview) tab with full REST API documentation for the AgentMark Gateway
  * Interactive endpoint docs for traces, scoring, templates, and health checks generated from an OpenAPI 3.1 spec
  * [Authentication guide](/api-reference/authentication) covering API key setup, request examples in curl, TypeScript, and Python, plus rate limit and quota details
  * New `agentmark api` CLI command for querying the gateway API directly from your terminal — auto-generated from the OpenAPI spec, targets local dev server by default or cloud with `--remote` *(both the CLI command and `--remote` were retired in CLI 0.13/0.14; use the [`agentmark-mcp`](/ai-editors/mcp-servers) MCP server or direct REST instead)*
  * **[Score analytics widgets](/observe/dashboards#score-analytics)** — four new widget types for [custom dashboards](/observe/dashboards#widgets): summary cards, score distribution histograms, trend-over-time charts, and cross-score comparison (confusion matrix or scatter plot). Start from the Score Analytics template or add individual widgets to any dashboard.
  * **[PII masking](/observe/pii-masking)** — redact sensitive data from traces before they leave your application. Available in both TypeScript and Python SDKs with a built-in masker for common patterns (email, phone, SSN, credit card, IP address), custom regex patterns, and zero-code environment variable suppression via `AGENTMARK_HIDE_INPUTS` / `AGENTMARK_HIDE_OUTPUTS`. Masking runs client-side with fail-closed behavior — no unmasked data ever reaches the network.
</Update>

<Update label="2026-04-10" description="Prompt Playground with Multi-Variant Comparison Mode">
  ### Features & Improvements

  * New [Playground](/build/playground) for comparing prompts across multiple models and configurations side-by-side
  * Run up to 6 variants simultaneously with independent model, temperature, and prompt override settings
  * Per-variant prompt editing with a "Modified" badge to highlight customized text
  * Output metadata chips showing model name, latency, token usage, and finish reason for each variant
  * One-click **Apply** to write the winning variant's configuration back to the editor
  * Responsive 3-column grid layout that scales from 2 to 6 variants
</Update>

<Update label="2026-04-10" description="Semantic Span Kinds, Observe Function, Span Kind Filters and Analytics">
  ### Features & Improvements

  * Expanded `SpanKind` enum from 3 to 7 values — added `agent`, `retrieval`, `embedding`, and `guardrail` span kinds for richer trace categorization
  * New `observe()` function for wrapping functions with automatic input/output capture and span kind tagging (available in both TypeScript and Python SDKs)
  * New "Span Kind" filter in the trace list — filter traces by `function`, `llm`, `tool`, `agent`, `retrieval`, `embedding`, or `guardrail`
  * New span kind breakdown analytics endpoint — view cost, latency, and token usage grouped by span type
  * Graph view now renders `embedding` and `guardrail` nodes with dedicated icons and colors
  * Cross-platform compatibility — the SDK now sets both `agentmark.span.kind` and `openinference.span.kind` attributes for portability with Langfuse and Phoenix

  ### Bug Fixes

  * Embedding spans are now correctly tracked for [cost analytics](/observe/cost-and-token-tracking) (previously excluded)
</Update>

<Update label="2026-03-09" description="Custom Roles & Permissions, Trace Search, Custom Dashboards, Billing Tiers, and More">
  ### Features & Improvements

  * Custom roles & granular permissions — create custom roles with fine-grained permission controls, prerequisite enforcement, and app-level role assignments
  * Permission-aware navigation — UI elements and pages automatically show/hide based on user permissions
  * Entitlement-gated permissions — permissions are automatically adjusted when billing plans change
  * [Trace search & saved filters](/observe/traces-and-logs) — search trace inputs/outputs, filter by span attributes, and save filter presets for quick access
  * Saved views with aggregate groupBy on the requests page
  * [Custom analytics dashboards](/observe/overview) — build personalized dashboards with configurable widgets
  * Eval score chips on trace tree span nodes for at-a-glance evaluation results
  * Rich hover tooltips on graph view nodes with metadata display
  * Trace I/O aggregation — generation span inputs/outputs surfaced at the trace root level
  * Streamlined publish UI — single-button publish flow with async confirmation dialog
  * Tiered billing plans — Team plan card, tier-aware checkout, and per-tier pricing
  * Tiered span limits and query-time data retention enforcement
  * Tiered feature gating system for plan-based access control
  * Alert commit SHA tracking — see which deployment triggered an alert
  * Automated model configuration sync — new models are auto-configured from the registry
  * Auto-generated editor schemas from the model registry
  * XML tag passthrough — lowercase XML tags now pass through in prompt content
  * Webhook secret masking for improved security
  * SSO SAML attribute mapping for user display names
  * Span usage notifications — improved delivery reliability
  * Billing confirmation dialog before plan upgrades
  * Trace tunnel — `agentmark dev` for one-step local-to-platform trace forwarding

  ### Bug Fixes

  * Fixed experiments page crash when the analytics backend is unavailable
  * Fixed trace view I/O tab not showing data for root traces
  * Fixed trace view tooltip crash on large JSON payloads
  * Fixed invite flow not persisting correctly for custom roles
  * Fixed profile visibility — tenant members can now view all profiles
  * Fixed MCP trace connection issues
  * Fixed deployment trigger button permission gating
  * Fixed score fetch errors silently failing in trace view
</Update>

<Update label="2026-02-04" description="Terms of Service, Admin Control Panel, Profile Updates, MCP API, and More">
  ### Features & Improvements

  * Terms of Service agreement — users are now prompted to accept terms before using the platform
  * Non-blocking terms update policy — users are notified of terms changes without being blocked
  * Streamlined registration — skip company setup during onboarding for faster account creation
  * Admin Control Panel — platform admins can manage users, roles, and organization settings
  * Profile email updates — update your profile email directly from account settings
  * Email sync — profile email now stays in sync when your authentication email changes
  * [MCP Server API endpoints](/ai-editors/mcp-servers) — programmatic access to the platform via MCP
  * Platform admin changelog notifications — send changelog updates to users via email
  * [Annotations](/evaluate/annotations) improvements and reliability fixes
  * Organization loading performance improvements

  ### Bug Fixes

  * Fixed OAuth registration flow
  * Fixed image prompt timeout for large image generation
  * Strengthened password validation with improved error messaging
  * Fixed request filters not working correctly
  * Fixed dashboard display issues
  * Fixed git sync and user invite flows
  * Fixed dataset run execution
  * Fixed GitLab commit history display
  * Fixed template frontmatter handling
</Update>

<Update label="2025-12-31" description="Multi-Tenant Membership, GitLab Integration, Tracing Revamp">
  ### Features & Improvements

  * Multi-tenant membership — users can now belong to multiple organizations and switch between them
  * GitLab integration — connect GitLab repositories alongside GitHub for prompt management

    [Read Docs](/build/version-control)
  * Tracing revamp — improved observability UI with better data presentation

    [Read Docs](/observe/traces-and-logs)

  ### Bug Fixes

  * Fixed trace span selection unexpectedly jumping to top span
  * Fixed edit prompt processing for 'done' message type
  * Fixed annotations not appearing after being added
</Update>

<Update label="2025-10-17" description="Annotations, Interactive Trace Graph, MCP Documentation">
  ### Features & Improvements

  * [Annotations](/evaluate/annotations) — add notes and labels to trace scores for better evaluation tracking
  * Interactive trace graph — visualize trace execution flow with resizable panels
  * [MCP integration documentation](/ai-editors/mcp-servers) — guides and examples for using AgentMark as an MCP server

  ### Bug Fixes

  * Fixed aspect ratio issue that could break the app layout
  * Fixed GitHub username not being saved on first OAuth login
  * Fixed webhook delivery reliability
</Update>

<Update label="2025-08-18" description="Evaluation Support, Dataset V2, Alerts, and More">
  ### Features & Improvements

  * [Evaluation support](/evaluate/writing-evals) — evaluate prompts and dataset outputs with scoring
  * [Eval-based alerts](/observe/alerts) — get notified when evaluations detect issues
  * [Dataset V2](/evaluate/datasets) — improved dataset management with file loading support
  * Enhanced dataset metrics and evaluation charts for better insights
  * Automated pricing updates — model pricing stays current automatically
  * Build failure notifications — project owners are notified when builds fail
  * Improved permission checks for server actions

  ### Bug Fixes

  * Fixed OAuth registration login and redirect issues
  * Fixed incorrect trace count when using filters
  * Improved database error messages with readable descriptions
</Update>

<Update label="2025-07-07" description="AgentMark init, CLI, Auth, Webhook, Streaming, Rollbacks, and more">
  ### Features & Improvements

  * AgentMark init
  * Updated examples in CLI init
  * Rebrand Puzzlet -> AgentMark
  * CLI: “run-prompt” for dataset + single props
  * Webhook Helpers
  * Alerts enhancements
  * Google/GitHub auth
  * Dataset runs directly via prompts
  * Vercel v4 webhook helper
  * Streaming to the platform
  * Commit History + Rollbacks

  ### Bug Fixes

  * Ollama fix on init
</Update>

<Update label="2025-06-08" description="Attachments, Dataset references, MCP Server, Editor UI, and more">
  ### Features & Improvements

  * File/Image Attachments
  * Dataset references in prompts
  * MCP Server on init
  * Popout editor to prompt input
  * Improved datasets UI in editor

  ### Bug Fixes

  * Disable publish button for read users
</Update>

<Update label="2025-05-09" description="AgentMark v3, Image/Speech Models, Vercel Adapter, Type-Safety, and more">
  ### Features & Improvements

  * AgentMark v3
  * Support Image + Speech Models
  * Vercel AI Adapter
  * Type-Safety for tools
  * Internal: Feature flag support
  * Commit messages

  ### Bug Fixes

  * General bug fixes
</Update>

<Update label="2025-04-09" description="JSONL Datasets, Evals/Scoring, and more">
  ### JSONL Datasets

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/jsonl.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=8f9e4dd10a1cfb978a5fb71058eed5c8" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="JSONL dataset file format in AgentMark for bulk testing with streaming support" width="1272" height="256" data-path="images/changelog/jsonl.png" />

  Datasets are now supported with JSONL files. This allows you to test your prompts in bulk against large datasets, and supports streaming.

  [Read Docs](/evaluate/datasets)

  ### Evals & Scoring

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/evals.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=799c24e1d31a38be2802eda10b85e23a" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="AgentMark Evals interface showing prompt evaluation scores and metrics" width="1456" height="840" data-path="images/changelog/evals.png" />

  AgentMark rolled out its initial evals support. Evals allow you to evaluate your prompts against a set of data, and get a score. More to come here soon.

  [Read Docs](/evaluate/writing-evals)

  ### Other

  * Consolidating prompts, evals, and datasets into single "files"
  * Officially rolled out alerts
  * Some CLI improvements
  * Minor bug fixes
</Update>

<Update label="2025-03-12" description="Sessions, Alerts, Trace UI Improvements, Onboarding Improvements">
  ### Sessions

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/E5mamgutKtLDWw1H/images/platform/observability/sessions.png?fit=max&auto=format&n=E5mamgutKtLDWw1H&q=85&s=25e9815e81ecc95f9817856dfd43391d" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Sessions view grouping related traces together for workflow debugging" width="1919" height="909" data-path="images/platform/observability/sessions.png" />

  Sessions provide a way to group related traces together, making it easier to monitor and debug complex workflows in your LLM applications. By organizing traces into sessions, you can track the entire lifecycle of a user interaction or a multi-step process.

  [Read Docs](/observe/sessions)

  ### Alerts

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/alerts.jpeg?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=5088cb1573496304025e239c8c80d88b" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Alerts configuration for error, latency, and cost monitoring via Slack or webhook" width="500" height="500" data-path="images/changelog/alerts.jpeg" />

  Now, you can get notified when your application is experiencing increased errors, latency, or costs. Configure alerts to notify you via Slack, or a webhook.

  [Read Docs](/observe/alerts)

  ### Traces UI Improvements

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/E5mamgutKtLDWw1H/images/platform/observability/traces.png?fit=max&auto=format&n=E5mamgutKtLDWw1H&q=85&s=5135ceec5e6390b18faff16c212dad4f" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Improved Traces UI displaying request details at a glance" width="1467" height="910" data-path="images/platform/observability/traces.png" />

  Traces now have a more user-friendly UI, with a focus on providing important information at a glance.

  ### Onboarding Improvements

  AgentMark improved its onboarding. Now, you can see your dashboard without having to sync your repo first. AgentMark also supports modular onboarding, so you can skip steps you don't need.
</Update>

<Update label="2025-02-18" description="Add Trace Examples to Datasets, Load Trace in Prompt, Re-indexing, App UI Improvements, bug fixes">
  ### Adding Examples to Datasets

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/add-to-dataset.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=39a60e529b60ad34f7e1c1f8dba29d8b" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="One-click button to add production trace data to datasets" width="1880" height="1210" data-path="images/changelog/add-to-dataset.png" />

  You can now add production trace data to your datasets with a single click.

  [Read Docs](/evaluate/datasets)

  ### Adding Examples to Prompts

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/add-to-dataset-prompt.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=8db508e208c63fcb0420a3de597a1e5b" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Adding production trace examples to prompts for testing with real data" width="626" height="216" data-path="images/changelog/add-to-dataset-prompt.png" />

  You can now add production trace examples to your prompts. This allows you to iterate/test against your prompts with real data.

  [Read Docs](/build/overview)

  ### Re-indexing

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/re-index.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=ad8eef6bfbf1a2d8beca17f9649e335f" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Re-index button to refresh prompts and datasets from synced repository" width="844" height="508" data-path="images/changelog/re-index.png" />

  You can now re-index your prompts, and datasets. This allows you to perform a fresh pull on the content from your synced repository.

  ### App UI Improvements

  You can now easily view your app's repo configuration, including repo names, branch, and more.
</Update>

<Update label="2025-01-27" description="Type Safety, Datasets, and more">
  ### Type Safety

  AgentMark aims to provide developers with the best developer experience possible. As part of this, type safety has just been added to the platform.

  * Types can now be generated via the CLI
  * Fetching prompts from the CDN or AgentMark are now type-safe
  * Prompts now support run/compile/deserialize functions

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/type-safety.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=f7599109c9c9d9e72c7bded824389d35" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Type-safe prompt fetching with TypeScript autocomplete in IDE" width="1278" height="1182" data-path="images/changelog/type-safety.png" />

  Read more about [Type Safety](/configure/type-safety)

  ### Datasets

  Datasets now allow you to test your prompts in bulk against a large set of data.

  * Run your datasets in bulk against your prompts
  * View previous runs, with inputs/outputs
  * View traces associated with each run
  * View high-level metrics for each run

  Read more about [Datasets](/evaluate/datasets)

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/datasets.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=dbc55960bbf0352d24cca319c957dbab" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Datasets dashboard showing bulk test runs with inputs, outputs, and metrics" width="2732" height="1422" data-path="images/changelog/datasets.png" />

  ### Trace Grouping

  Traces can now be grouped based on the trace function, and the component function. Trace groups together at the root level,
  while component allows for sub-groups.

  * New function added: `trace`
  * New function added: `component`

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/changelog/traces-components.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=4a2fa017e122689dc80986de20a5ca40" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" alt="Trace grouping with trace and component functions for hierarchical organization" width="902" height="862" data-path="images/changelog/traces-components.png" />

  ### CLI Improvements

  The CLI has been improved to provide a better developer experience.

  * AgentMark `init` can optionally create an example app
  * Added `pull-models` to walk through adding new models to your platform

  Read More about the [CLI](/getting-started/quickstart)

  ### Bug Fixes

  * Fixed a bug which could cause an app's templates to be deleted when a new app was created
  * Fixed a bug which could cause some branches not to show up in the UI
  * Fixed a bug which could prevent newly created local prompts from being synced to the platform

  ### Other

  * Improved UI for prompts input/output
  * Paginate traces
  * Improved UI theme for prompts
</Update>

<Update label="2025-01-16" description="Prompt Management, Observability, Datasets, CLI, Platform Management, and Evals">
  ### Overview

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/AVLe-j9jYV0ZQWXl/images/main/puzzle-backdrop.svg?fit=max&auto=format&n=AVLe-j9jYV0ZQWXl&q=85&s=840dfcca782a472d7649150a3dfda507" alt="AgentMark logo - git-based Prompt Engineering Platform" width="500" height="500" data-path="images/main/puzzle-backdrop.svg" />

  AgentMark is a git-based Prompt Engineering Platform where application developers and prompt engineers collaborate on GenAI products. AgentMark enables application developers to manage their configuration, prompts, datasets, and evals in a git-based workflow while also providing a hosted platform for collaboration with non-technical team members.

  ### Features

  * Prompt Management
  * Observability
  * Datasets
  * CLI
  * Platform Management
  * Evals

  ### Prompt Management

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/YfsTGlXrX2VdPtgF/images/dashboard/prompts.png?fit=max&auto=format&n=YfsTGlXrX2VdPtgF&q=85&s=9ab556ff92e6f4cf172148b73626c79b" alt="AgentMark Prompt Management dashboard showing prompts organized as files in a git-synced repository" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" width="2788" height="1594" data-path="images/dashboard/prompts.png" />

  AgentMark takes a developer-first approach to prompt management, treating prompts as files that live in your repository while still providing a platform for non-technical team members. All prompts are saved in AgentMark, a markdown-based format that's easy to write and read.

  [Read Docs](/build/overview)

  ### Observability

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/KmsdEn15yStMq4xE/images/platform/observability/dashboard.png?fit=max&auto=format&n=KmsdEn15yStMq4xE&q=85&s=8acd475d7befd16f40d42c027452ebff" alt="AgentMark Observability dashboard displaying LLM application metrics, traces, and logs" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" width="2382" height="1592" data-path="images/platform/observability/dashboard.png" />

  AgentMark builds on top of OpenTelemetry for collecting telemetry data from your prompts. This helps you monitor, debug, and optimize your LLM applications in production. AgentMark provides traces, logs, metrics, and more.

  [Read Docs](/observe/overview)

  ### Datasets

  Create datasets to easily test your prompts in bulk against a large set of data.

  [Read Docs](/evaluate/datasets)

  ### CLI

  AgentMark provides a CLI for initializing your AgentMark app, customizing it, and deploying it to the cloud. Add new models to your platform with just a single command.
  You can also develop w/ AgentMark locally using the serve command.

  ```bash theme={null}
  npx create-agentmark
  ```

  [Read Docs](/getting-started/quickstart)

  ### Platform Management

  AgentMark offers an intuitive platform for creating new git-synced apps, adding team members with roles, and setting up API keys for users.

  ### AgentMark SDK

  <img src="https://mintcdn.com/puzzlet-9ba7bb98/E5mamgutKtLDWw1H/images/platform/observability/traces.png?fit=max&auto=format&n=E5mamgutKtLDWw1H&q=85&s=5135ceec5e6390b18faff16c212dad4f" alt="AgentMark SDK traces view showing LLM request spans with timing and metadata" className="w-full rounded-xl border border-gray-800 shadow-2xl mb-12" width="1467" height="910" data-path="images/platform/observability/traces.png" />

  AgentMark's SDK is simple and easy to use. It offers features like: one-LOC observability, securely fetching prompts from the CDN, and more.

  [Read Docs](/introduction/core-concepts)
</Update>

<Update label="2025-01-03" description="Initial AgentMark Release">
  ### Features

  * Initial release of AgentMark
  * Support for OpenAI, Anthropic, and other LLM providers
  * MDX-based prompt templating
  * Type-safe prompt development
  * Tools and agents support

  ### Documentation

  * Added comprehensive documentation
  * Included examples and guides
  * API reference documentation
</Update>