Skip to content
Talk to Us

Blocks.ai publishes this comparison and sells in this market. Every claim about another product links to a public source.

LangSmith vs Zenity

LangSmith

Platform for observing, evaluating, and deploying agents

Zenity AI Agent Security & Governance Platform

Security and governance platform for AI agents, aimed at security teams

Short answer

LangChain calls LangSmith a platform for observing, evaluating, and deploying agents; Zenity says it is an AI agent security and governance platform for security teams.⁠Source 1, Source 2, Source 3 LangSmith runs agents and builds no-code ones in Fleet; Zenity says it catalogs agents and can block actions on Copilot Studio, Microsoft Foundry, and coding agents.⁠Source 4, Source 5, Source 6, Source 7

By Blocks.ai teamLast checked

Where each one sits

Six layers of running AI agents at a company, and what each product’s own public sources say it covers.

These aren’t the same kind of product

Where they overlap
Both keep records of agent activity: LangSmith creates a tracing project per deployment and traces Fleet agents’ tool calls; Zenity says it logs agent messages, tool calls, retrievals, and handoffs.⁠Source 6, Source 8, Source 9
Where they differ
LangSmith builds and runs agents.⁠Source 4, Source 5 Zenity says it inventories and governs agents that run on other platforms, such as Copilot Studio and Agentforce.⁠Source 6, Source 10
Running both
Zenity says any agent that emits OpenTelemetry or gen_ai spans can connect to it.⁠Source 11 Neither vendor publicly documents using the two together.
Public sources · checked 2 October 2026
  • Offered
  • Not publicly documented

LangSmith

  • Build agents: OfferedFleet agent builder⁠Source 5
  • Host and run agents: OfferedLangSmith Deployment runtime⁠Source 4
  • Identity and access: OfferedFleet per-agent permissions⁠Source 12
  • Registry and governance: OfferedCentralized registry in Deployment⁠Source 13
  • Traffic between agents, tools, and models: OfferedFleet forwards MCP tool calls⁠Source 14
  • Agents across organizations: Not publicly documented

Zenity

  • Build agents: Not publicly documented
  • Host and run agents: Not publicly documented
  • Identity and access: Not publicly documented
  • Registry and governance: OfferedAI Observability agent inventory⁠Source 6
  • Traffic between agents, tools, and models: OfferedTool-call blocking for coding agents⁠Source 15
  • Agents across organizations: Not publicly documented

At a glance

TopicLangSmithZenity
What it isLangChain calls it an agent engineering platform for observing, evaluating, and deploying agents, with Fleet for non-technical teams.⁠Source 1A security and governance platform for AI agents, which Zenity aims at security teams that need to discover and inventory agents and enforce policies.⁠Source 2, Source 3
What it doesRuns agents in production on the Plus plan or above, builds agents without code in Fleet, and traces agent activity.⁠Source 4, Source 5, Source 8, Source 9, Source 16Zenity says it finds agents in your environment, logs agent activity, and can check agent actions against your rules.⁠Source 6, Source 17
Agents it coversAgents deployed on LangSmith, built with LangChain or other frameworks such as Google ADK and CrewAI.⁠Source 18 Agents hosted elsewhere can send it traces.⁠Source 19Zenity says it covers agents in Microsoft Copilot Studio, ChatGPT Enterprise, and Salesforce Agentforce, homegrown agents, and coding agents such as Cursor.⁠Source 6, Source 15
Where it runsLangChain’s cloud, on GCP or AWS.⁠Source 20 On Enterprise, also Bring Your Own Cloud on AWS, self-hosted, or hybrid.⁠Source 19, Source 21, Source 22Software as a service, shown on AWS Marketplace as deployed on AWS.⁠Source 2 Self-hosting isn’t in Zenity’s public docs; its product docs require a login.⁠Source 23
Pricing modelDeveloper includes one free seat; Plus is $39 per seat a month; usage is metered in LangChain Standard Units (LSUs), $1 each.⁠Source 22 For Enterprise, contact sales.⁠Source 22Custom pricing on its main AWS Marketplace listing; a Security Hub Extended listing shows usage prices per resource and per million tokens.⁠Source 2, Source 24

What each one is

LangSmith

LangChain describes LangSmith as a framework-agnostic platform for observing, evaluating, and deploying agents.⁠Source 1 Its Deployment service is a runtime for agents in production, and its Fleet service is for creating and managing agents without code.⁠Source 4, Source 5 Workspace role-based access control and audit logs are Enterprise features.⁠Source 25, Source 26

Zenity AI Agent Security & Governance Platform

Zenity is a security and governance platform for AI agents, spanning SaaS, homegrown cloud platforms, and end-user devices.⁠Source 2 Zenity says AI Observability catalogs agents and Runtime Boundaries can check agent actions against your rules.⁠Source 6, Source 17 It is delivered as software as a service.⁠Source 2

The differences that matter

  1. Which agents each one covers

    LangSmith

    LangSmith deploys agents built with LangChain or other frameworks, such as Google ADK and CrewAI; agents hosted elsewhere can send it traces.⁠Source 18, Source 19

    Zenity

    Zenity says it catalogs agents where they already run: inside SaaS platforms such as Copilot Studio and Agentforce, custom-built, or on laptops.⁠Source 6

    Listing agents not deployed or built in LangSmith is not publicly documented; building agents in Zenity isn’t in Zenity’s public docs.

  2. Controlling what agents do

    LangSmith

    Fleet agents can pause for approval before specific actions, and per-agent permissions set who can clone, run, or edit each one.⁠Source 9, Source 12

    Zenity

    Runtime Boundaries can check agent actions in real time; Zenity says it can block actions only on Copilot Studio, Microsoft Foundry, and coding agents.⁠Source 7, Source 17

    On Enterprise, Fleet admins can grant or restrict users’ access to MCP servers and integrations by tag.⁠Source 9 Zenity says its posture policies can alert, block, or automatically remediate.⁠Source 10

  3. Where it runs

    LangSmith

    LangSmith Cloud is hosted on GCP or AWS; on the Enterprise plan, LangSmith can be self-hosted or run as Bring Your Own Cloud on AWS.⁠Source 19, Source 20, Source 21

    Zenity

    Zenity is delivered as software as a service, deployed on AWS per its AWS Marketplace listing.⁠Source 2

    LangSmith Cloud stores the resources and services its docs list in your sign-up region.⁠Source 20 Zenity’s hosting regions, data residency, and self-hosting aren’t in Zenity’s public docs.

For security teams

What a security review asks, answered from each vendor’s public documentation.

TopicLangSmithZenity
Network exposureEnterprise can reach LangSmith Cloud privately.⁠Source 20 On self-hosted LangSmith with Deployment, agents run behind an ingress you configure.⁠Source 27Delivered as SaaS on AWS, per its AWS Marketplace listing.⁠Source 2 Monitored agents’ network needs aren’t in Zenity’s public docs.
IdentityHosted Agent Servers default to API keys; self-hosted ones have no default authentication.⁠Source 28 Cloud Enterprise adds SAML SSO and SCIM.⁠Source 29Zenity says Boundaries rules can use Okta attributes such as role.⁠Source 17 Issuing agent identities isn’t in Zenity’s public docs.
Access changes and revocationFleet sharing can be revoked at any time; organization admins can deactivate or delete any member’s personal access token.⁠Source 9, Source 30Zenity says its kill switch immediately disables an agent’s tool and data access.⁠Source 17 It says rules can shut agents down.⁠Source 17
Audit trailEnterprise audit logs: admin and configuration actions, kept up to 400 days.⁠Source 26 You can pull them into a SIEM hourly.⁠Source 26Zenity says it logs agent messages and tool calls, and can stream audit log events to Splunk or Microsoft Sentinel.⁠Source 6, Source 11
ComplianceLangChain’s Trust Center lists SOC 2 Type II, ISO 27001:2022, GDPR, and HIPAA for LangSmith.⁠Source 31Zenity says it has SOC 2 Type II, is ISO 27001 compliant, and announced FedRAMP “In Process” in March 2026.⁠Source 32, Source 33

Full comparison

18 criteria in five groups. Every cell links to its source, or says no public source answers it.

LangSmith and Zenity compared on 18 criteria
LangSmithZenity
What it is
What it is and who it’s forLangChain calls it a framework-agnostic platform for observing, evaluating, and deploying agents.⁠Source 1 Deployment runs agents; Fleet builds agents without code.⁠Source 4, Source 5A SaaS security and governance platform for AI agents spanning SaaS, homegrown cloud platforms, and end-user devices, which Zenity aims at security teams.⁠Source 2, Source 3
MaturityLangChain announced GA of LangGraph Platform (now Deployment) on 14 May 2025 and Agent Builder (now Fleet) on 13 January 2026.⁠Source 13, Source 34, Source 35, Source 36Zenity announced Azure Marketplace (now Microsoft Marketplace) availability in March 2025 and AWS in January 2026.⁠Source 37, Source 38, Source 39
Control
Agent registry and discoveryLangChain says Deployment manages agents in a centralized registry with versioning and rollbacks.⁠Source 13, Source 40 Listing agents hosted elsewhere is not publicly documented.Zenity says AI Observability scans and catalogs agents in SaaS platforms, custom builds, and on laptops, with their permissions and tool access.⁠Source 6
Identity and access controlHosted Agent Servers use LangSmith API keys by default, or your own auth handler.⁠Source 28 Enterprise adds RBAC, plus SAML SSO and SCIM on Cloud.⁠Source 25, Source 29Zenity says Boundaries rules can use Okta attributes such as active status, role, and job title.⁠Source 17 Issuing agent identities isn’t in Zenity’s public docs.
Ownership, policy, and revocationFleet shows each agent’s owner and has per-agent Clone, Run, and Edit permissions, revocable at any time; agents can be set to pause for approval.⁠Source 9, Source 12, Source 36Zenity says rules can allow an action, block it, or shut the agent down, but it blocks inline only on Copilot Studio, Microsoft Foundry, and coding agents.⁠Source 7, Source 17
Audit log and observabilityEnterprise audit logs of admin and configuration actions, kept up to 400 days.⁠Source 26 Each deployment gets a tracing project; Fleet traces include tool calls.⁠Source 8, Source 9Zenity says it logs agent messages, tool calls, retrievals, and handoffs.⁠Source 6 It says its audit log events can stream to Splunk or Microsoft Sentinel.⁠Source 11
Connection
How agents connectOn self-hosted LangSmith with Deployment, agents run behind an ingress you configure.⁠Source 27 Enterprise customers can reach LangSmith Cloud over AWS PrivateLink.⁠Source 20Zenity says agents emitting OpenTelemetry or gen_ai spans can connect, with an Evaluate API for inline enforcement.⁠Source 11 Network needs: not in Zenity’s public docs.
Agents across organizationsAgent Server agents can communicate with other A2A-compatible agents.⁠Source 41 An access model for other organizations’ agents is not publicly documented.Not in Zenity’s public docs; its product docs require a login (checked 2 October 2026)⁠Source 23
Protocol supportAgent Server supports A2A over JSON-RPC, with documented gaps, and can expose agents as MCP tools.⁠Source 41, Source 42 Fleet discovers tools from remote MCP servers.⁠Source 14Zenity says custom agents can connect by emitting OpenTelemetry or gen_ai spans.⁠Source 11 Whether Zenity supports A2A isn’t publicly documented.
Frameworks, models, and clouds supportedDeployment runs agents from other frameworks, such as Google ADK, CrewAI, and AutoGen.⁠Source 18 Agents hosted elsewhere can send traces to LangSmith.⁠Source 19Zenity says it covers agents in Copilot Studio and Agentforce, homegrown agents on Bedrock or Vertex AI, any agent emitting OpenTelemetry, and coding agents.⁠Source 6, Source 11, Source 15
Operations
Deployment options and data residencyCloud on GCP (US, EU, APAC) or AWS (US).⁠Source 20 BYOC on AWS, or self-hosted or hybrid, on Enterprise.⁠Source 19, Source 21, Source 22Software as a service, shown on AWS Marketplace as deployed on AWS.⁠Source 2 Regions, data residency, and self-hosting aren’t in Zenity’s public docs.
Compliance attestationsLangChain’s Trust Center lists SOC 2 Type II, ISO 27001:2022, GDPR, and HIPAA for LangSmith.⁠Source 31Zenity says the company holds SOC 2 Type II certification and is ISO 27001 compliant.⁠Source 32 It announced FedRAMP “In Process” status in March 2026.⁠Source 33
Support and SLABase, Standard, and Premium support plans.⁠Source 43 Reasonable commercial efforts toward 99.5% API uptime for SaaS, measured quarterly.⁠Source 43Zenity’s terms commit to at least 99.9% monthly uptime, with commercially reasonable efforts.⁠Source 44 Support requests go through Zendesk in business hours.⁠Source 44
Time and effort to get runningCloud deployment needs the Plus plan or above and an API key.⁠Source 45 Fleet agents start from a description or a template.⁠Source 5Zenity says it has custom-agent guides for Cribl, LiteLLM, and Kong, and a Cursor plugin.⁠Source 11, Source 46 Prerequisites aren’t in Zenity’s public docs.
Pricing model and public pricesDeveloper includes one free seat; Plus is $39 per seat a month.⁠Source 22 Usage is metered in LSUs, $1 each.⁠Source 22 Enterprise: contact sales.⁠Source 22Main AWS listing: custom pricing.⁠Source 2 Security Hub Extended listing: Observability $130 per resource a month, Runtime Protection $16 per million tokens a month.⁠Source 24
Building
Agent building toolsFleet builds agents without code from a description or template.⁠Source 5 Code-first agents can be deployed with the LangGraph CLI and app templates.⁠Source 4Not in Zenity’s public docs; its product docs require a login (checked 2 October 2026)⁠Source 23
Model accessLangChain says LangSmith works with closed or open models.⁠Source 1 Fleet’s managed picker has three tiers (Fast, Pro, Max), not custom models.⁠Source 47Zenity says its threat detection combines rules mapped to OWASP LLM and MITRE ATLAS with LLM-based detections.⁠Source 48 The models used aren’t in Zenity’s public docs.
Integrations and ecosystemFleet has built-in tools such as Gmail, Slack, Google Calendar, and GitHub, and a central tool registry for admins.⁠Source 36, Source 47Zenity says its signals can show in Microsoft Agent 365 and findings in AWS Security Hub Extended; it is on the Cursor Marketplace.⁠Source 46, Source 49, Source 50

Which to choose

Choose LangSmith if

  • You want a managed runtime for agents built with LangChain or other frameworks, such as Google ADK, CrewAI, or AutoGen.⁠Source 4, Source 18
  • You want non-technical teams to build agents without code in Fleet, with per-agent Clone, Run, and Edit permissions.⁠Source 1, Source 5, Source 12
  • You need LangSmith’s control plane and Agent Servers in your own infrastructure, for full data residency or air-gapped use, on Enterprise.⁠Source 4
  • You want what LangChain calls a framework-agnostic agent engineering platform for tracing and evaluating agents.⁠Source 1

Choose Zenity if

  • Your security team needs one inventory of agents across platforms such as Copilot Studio and Agentforce, which Zenity says it builds.⁠Source 3, Source 6
  • You want the rule checks Zenity says it runs on agent actions, blocking inline on Copilot Studio, Microsoft Foundry, and coding agents.⁠Source 7, Source 17
  • You need the boundaries Zenity says it enforces on coding agents, such as Claude Code, GitHub Copilot, and Cursor.⁠Source 15
  • You want what Zenity says it offers: audit log events in Splunk or Microsoft Sentinel, or findings in AWS Security Hub Extended.⁠Source 11, Source 50

Questions buyers ask

How is each one priced?

LangSmith’s Developer plan includes one free seat, Plus is $39 per seat a month, and usage is metered in LSUs, $1 each.⁠Source 22 Zenity’s main AWS listing is custom-priced; a Security Hub Extended listing shows usage prices, such as $130 per resource a month for Observability.⁠Source 2, Source 24

Do LangSmith and Zenity support MCP and A2A?

LangSmith’s Agent Server supports A2A, with documented gaps, and can expose agents as MCP tools.⁠Source 41, Source 42 For coding agents, Zenity says it can block or modify a dangerous tool call through native agent hooks.⁠Source 15 Whether Zenity supports A2A isn’t publicly documented.

Can either one connect agents across organizations?

LangSmith’s Agent Server lets agents communicate with other A2A-compatible agents.⁠Source 41 An access model for agents owned by another organization is not publicly documented for either one.

Can either one stop an agent?

In Fleet, you pause an agent by pausing its channels, and a LangSmith Cloud deployment can be deleted.⁠Source 51, Source 52 Zenity says its Runtime Boundaries can shut an agent down.⁠Source 17 Zenity says its kill switch immediately disables an agent’s tool and data access.⁠Source 17

How we compare

Read the full method

Every claim on this page links to a public source. Where none answers a question, the page says so.

We re-check every fact at least every 90 days. This page was last checked .

Something wrong or out of date? Tell us and we’ll correct it.

Sources

57 public sources, each with the date we checked it. Every one opens in a new tab.

  1. Source 1: LangSmith: The Agent Engineering Platform LangChain · checked Back:abcdefg

  2. Source 2: AWS Marketplace: Zenity Zenity · checked Back:abcdefghijkl

  3. Source 3: Platform | AI Agent Security & Governance Platform | Zenity Zenity · checked Back:abcd

  4. Source 4: LangSmith Deployment LangChain · checked Back:abcdefghi

  5. Source 5: No-code agents with LangSmith Fleet LangChain · checked Back:abcdefghi

  6. Source 6: AI Observability | See Every Agent, Know What it Touches | Zenity Zenity · checked Back:abcdefghijklm

  7. Source 7: Zenity's Coverage of the 2026 OWASP Top 10 for LLM Apps Zenity · checked Back:abcd

  8. Source 8: LangSmith control plane LangChain · checked Back:abc

  9. Source 9: Access & oversight LangChain · checked Back:abcdefg

  10. Source 10: AI Security Posture Management (AISPM) | Stop Agent Risk Before Deployment | Zenity Zenity · checked Back:ab

  11. Source 11: From Triage to Full Coverage: The Shift AI Agent Security Took in August Zenity · checked Back:abcdefgh

  12. Source 12: Agent platform comparison LangChain · checked Back:abcd

  13. Source 13: LangSmith Deployment LangChain · checked Back:abc

  14. Source 14: Remote MCP servers LangChain · checked Back:ab

  15. Source 15: Coding and Personal Agents | Zenity Zenity · checked Back:abcde

  16. Source 16: Trace with OpenTelemetry LangChain · checked Back to text

  17. Source 17: Runtime Boundaries | The Runtime Boundary for Autonomous AI | Zenity Zenity · checked Back:abcdefghijk

  18. Source 18: Deploy other frameworks LangChain · checked Back:abcd

  19. Source 19: Set up LangSmith LangChain · checked Back:abcdef

  20. Source 20: Cloud (SaaS) LangChain · checked Back:abcdef

  21. Source 21: Bring Your Own Cloud (BYOC) LangChain · checked Back:abc

  22. Source 22: LangSmith Plans and Pricing LangChain · checked Back:abcdefgh

  23. Source 23: Zenity Documentation (login) Zenity · checked Back:abc

  24. Source 24: AWS Marketplace: Zenity AI Security & Governance Platform for Security Hub Extended Zenity · checked Back:abc

  25. Source 25: Role-based access control LangChain · checked Back:ab

  26. Source 26: Audit logs LangChain · checked Back:abcd

  27. Source 27: Enable additional LangSmith features LangChain · checked Back:ab

  28. Source 28: Authentication & access control LangChain · checked Back:ab

  29. Source 29: Authentication methods LangChain · checked Back:ab

  30. Source 30: Overview (administration) LangChain · checked Back to text

  31. Source 31: Trust Center - LangChain LangChain · checked Back:ab

  32. Source 32: Zenity Trust Center Zenity · checked Back:ab

  33. Source 33: Zenity Achieves FedRAMP “In Process” Status for AI Agent Security Zenity · checked Back:ab

  34. Source 34: LangGraph Platform is now Generally Available: Deploy & manage long-running, stateful Agents LangChain · checked Back to text

  35. Source 35: Now GA: LangSmith Agent Builder LangChain · checked Back to text

  36. Source 36: LangSmith Fleet changelog LangChain · checked Back:abc

  37. Source 37: Zenity Now Available in the Microsoft Azure Marketplace Zenity · checked Back to text

  38. Source 38: Introducing Microsoft Marketplace - Thousands of solutions. Millions of customers. One Marketplace. - The Official Microsoft Blog Zenity · checked Back to text

  39. Source 39: Zenity Now Available on AWS Marketplace, Bringing End-to-End Security to Amazon Bedrock AgentCore and Enterprise AI Agents Everywhere Zenity · checked Back to text

  40. Source 40: Assistants LangChain · checked Back to text

  41. Source 41: A2A endpoint in Agent Server LangChain · checked Back:abcd

  42. Source 42: MCP endpoint in Agent Server LangChain · checked Back:ab

  43. Source 43: Support Plans LangChain · checked Back:ab

  44. Source 44: Zenity Subscription Terms and Conditions (EULA linked from Zenity's AWS Marketplace listings) Zenity · checked Back:ab

  45. Source 45: Deploy your app to cloud LangChain · checked Back to text

  46. Source 46: Seeing Every MCP Connection: Zenity Joins the Cursor Marketplace Zenity · checked Back:ab

  47. Source 47: Essentials LangChain · checked Back:ab

  48. Source 48: AI Detection and Response (AIDR) | See the Threat, Stop the Action | Zenity Zenity · checked Back to text

  49. Source 49: Zenity Now Integrates with Microsoft Agent 365 Zenity · checked Back to text

  50. Source 50: Zenity Selected for AWS Security Hub Extended to Secure Enterprise AI Agents Zenity · checked Back:ab

  51. Source 51: Manage agent settings LangChain · checked Back to text

  52. Source 52: Manage a deployment LangChain · checked Back to text

  53. Source 53: Your company's private network Blocks.ai · checked Back to text

  54. Source 54: Network requirements Blocks.ai · checked Back to text

  55. Source 55: Solutions: Agent sprawl Blocks.ai · checked Back to text

  56. Source 56: Solutions: Partner networks Blocks.ai · checked Back to text

  57. Source 57: Pricing Blocks.ai · checked Back to text