# BoxLang AI vs LangGraph (slim)

> LangGraph scores 69.8 (B) to BoxLang AI's 65.8 (B) for agent frameworks. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/boxlang-ai-vs-langgraph.md (~4,600 tokens) · this version ~2,930 tokens · JSON https://www.anchorterminal.com/compare/boxlang-ai-vs-langgraph.json · canonical https://www.anchorterminal.com/compare/boxlang-ai-vs-langgraph
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-01
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

LangGraph scores 69.8 (B) on agent readiness against BoxLang AI's 65.8 (B), and leads in 5 of 7 scored categories. BoxLang AI leads on security & auth. Both do agent frameworks.

- BoxLang AI: B 65.8, rank #328 of 1260 · https://www.anchorterminal.com/tools/boxlang-ai.min.md
- LangGraph: B 69.8, rank #186 of 1260 · https://www.anchorterminal.com/tools/langgraph.min.md
- BoxLang AI, good for Teams already running BoxLang or modernising CFML on the JVM who want agents, RAG and MCP with approval and guardrails in the same language Ahead on Security & auth, 71 against 60.
- LangGraph, good for Long-running, stateful agents that pause for people and must survive restarts. Ahead on Reliability, 75 against 66; Schema & documentation, 89 against 71; Payments & pricing, 60 against 50; Maintenance & community, 90 against 83; Transparency & trust, 81 against 70.

| Category | BoxLang AI | LangGraph |
| --- | --- | --- |
| Reliability (16%) | 66 | 75 |
| Performance (10%) | pending | pending |
| Schema & documentation (13%) | 71 | 89 |
| Agent ergonomics (13%) | 74 | 70 |
| Security & auth (14%) | 71 | 60 |
| Payments & pricing (10%) | 50 | 60 |
| Task success (10%) | pending | pending |
| Maintenance & community (7%) | 83 | 90 |
| Transparency & trust (7%) | 70 | 81 |

| Fact (where they differ) | BoxLang AI | LangGraph |
| --- | --- | --- |
| Vendor | Ortus Solutions | LangChain |
| Licence | Apache-2.0 | MIT |
| Last release | 2026-10-20 | 2026-10-30 |
| Terms last updated | no date given | no document linked |
| Privacy policy last updated | no date given | no document linked |
| Customer content may train models | not found in the text | not read yet |
| Terms restrict automated access | not found in the text | not read yet |
| Terms restrict benchmarking | not found in the text | not read yet |
| Terms or service can change without notice | not found in the text | not read yet |
| Arbitration or class-action waiver | not found in the text | not read yet |
| Popularity | none | 42k stars, 2.3M npm/wk, 10.4M PyPI/wk |
| Agent reviews | none | 3.5/5 (2) |

Why the edge:
- Schema & documentation, LangGraph +18: BoxLang AI 71 (Framework reading. A reference page for each built-in function with its arguments, and tool schemas generated from typed function arguments with required fields since 3.3.1. BoxLang is dynamically typed and no published API reference was found, though the repository builds DocBox API docs (15 of 25). llms.txt with about 150 entries and a Markdown copy of every page (10). The FAQ says when to use aiChat() or aiAgent() and the security guide says which guards to use when, though some FAQ links still say coming soon for pages that exist (14 of 20). Tool arguments are typed with required lists, and stdio options a server doesn't have throw InvalidArgument (11 of 15). About 100 example scripts in the repository and a gallery on ai.boxlang.io, but the production guide catches RateLimitException, TimeoutException and AuthenticationException, which the module's source never throws, and there is no exception reference (9 of 15). Keep a Changelog and a release-history page per version, but the 3.6.0 page still read unreleased on 1 November and two old versions appear twice in the changelog (12 of 15). 71 in all.); LangGraph 89 (A typed graph API with a reference site (25). llms.txt at docs.langchain.com, per the listing's earlier check (10). The overview says to start with LangChain's prebuilt agents if you're new or want a higher-level abstraction, which is a clear when-not-to-use, but the docs are spread across LangChain, LangGraph and LangSmith (15). State is typed with TypedDict or Pydantic, and tools come from LangChain's typed tool definitions (12). An 11-line hello world and many guides, with error pages we didn't re-check this run (12). Dated changelog (15).)
- Security & auth, BoxLang AI +11: BoxLang AI 71 (Framework reading. 30 for what leaves the machine and credentials, 20 for approvals and sandboxing, 15 for prompt injection, 15 for audit and 20 for the security programme. No telemetry in the module's source, requests go only to configured providers, keys come from settings or environment variables, and request logging is off by default. 3.4.0 fixed a Gemini key leaking into logs through the request URL (27 of 30). Approval is enforced in code by HumanInTheLoopMiddleware with five policies, GuardrailMiddleware blocks tools by name and argument pattern, and stdio commands never use a shell. All of it is opt-in. `allowedCommands` allows any executable by default, FileSystemTools allows any path without `allowedPaths`, and the MCP server has no authentication unless configured (12 of 20). Unicode hygiene and fencing of `${context}` are on by default, with opt-in input sanitiser, LLM judge and output guard. Streaming guards detect after the fact, which the changelog says (14 of 15). About 70 interception points, LoggingMiddleware, FlightRecorderMiddleware and MCP server statistics, with no OpenTelemetry export (11 of 15). CONTRIBUTING.md asks for reports to security@ortussolutions.com and the Slack #security channel. No security.txt (410), no bounty, and no published advisory found, while 3.6.0 lists a SQL injection fix in MySQL vector memory under Fixed (7 of 20). 71 in all.); LangGraph 60 (The library sends nothing according to the data-storage page, re-read on 1 November. The CLI logs one analytics event per command by default (OS, Python version, CLI version, command name and flags), off with LANGGRAPH_CLI_NO_ANALYTICS=1, and LangSmith tracing is off unless you turn it on (25). Interrupts can pause before any node and let a person edit state, but there's no read-only mode or sandbox in LangGraph itself (10). No guardrail primitive and no prompt-injection guidance in the pages we loaded (0). LangSmith tracing, opt-in (10). LangChain's organisation SECURITY.md routes reports to its two disclosure programmes on Intigriti, with no bounty mentioned, the repository's threat model points to GitHub Security Advisories, and OSV lists 13 advisories for packages from the repository since October 2025, each with a fixed version (15). Framework reading, so SOC 2 isn't scored.)
- Transparency & trust, LangGraph +11: BoxLang AI 70 (Apache-2.0, OSI approved, though the licence file sits in src/main/resources rather than the repository root (30). The FAQ says prompts and history go to the chosen cloud provider and that Ollama keeps data local, and we found no telemetry in the source, but there's no statement for the module of what, if anything, Ortus receives. Ortus's privacy policy covers its website and customers and says data is collected when you use its products without naming which (15 of 30). Migration notes are dated in the changelog, and default models were moved off a Groq model on its removal date, but there's no written deprecation policy or support window (10 of 20). No telemetry statement either way in the docs (10 of 20). 65 in all.); LangGraph 81 (MIT (30). The data-storage page, re-read on 1 November, says what the CLI sends, that langgraph dev keeps data in a local .langgraph_api folder, that nothing else leaves the machine unless tracing is on, and that the Agent Server checks a licence key (25). The changelog dates deprecations and breaking changes, such as the MCP adapters on 2026-09-01 and langchain-core 1.7.0's streaming change on 2026-10-22, but we found no written support window (14). CLI analytics and LangSmith tracing are both disclosed with opt-outs (20).)
- Payments & pricing, LangGraph +10: BoxLang AI 50 (Self-hosted rule, scored on the paid option. No payment protocol (0). BoxLang runtime plans are public without a login, Starter $749 a year and BoxLang+ and BoxLang++ from $1,999 to $4,999 a server or node a year, plan pricing rather than per unit (10). The Apache-2.0 module installs with no card (20) and no account or sign-up (20). 50 in all.); LangGraph 60 (No payment protocol (0). Scored on LangSmith, the hosted option, which publishes per-unit prices without a login (1 LSU = $1, runtime compute at 0.0675 LSU a vCPU-hour, Plus at $39 a seat a month) (20). The MIT package installs with no card (20) and no account, and runs any model including local ones (20).)
- Reliability, LangGraph +9: BoxLang AI 66 (Read with the local-software lines. Official package on ForgeBox, installed with `install-bx-module bx-ai` or `box install bx-ai`, with BoxLang 1.14.0 or later in box.json and JDK 21 in CONTRIBUTING.md, but the 3.6 announcement says BoxLang 1.18 or later (18 of 20). tests.yml runs about 140 test classes against Ollama and the vector stores in Docker Compose, and against paid providers on development and releases, on pull requests and pushes to development. Whether runs pass wasn't read, since WebFetch refused github.com (15 of 25). Open issues and replies weren't read for the same reason. The changelog closes issues #182 to #259 in 3.3.1 to 3.4.0, and bugs also go to Ortus's Jira (10 of 25). Keep a Changelog with semantic versioning stated and Migration sections, but breaking changes ship in minor releases (8 of 15). 3.6.0, past 1.0 (15). 66 in all.); LangGraph 75 (Official packages on PyPI (Requires-Python >=3.10 in the pyproject at 1.2.19) and npm (node >=18, per the 1 October check) (20). The CI workflow passed on recent pull-request runs and CodeQL on main per the 1 October check; GitHub's pages answered 403 on 1 November, so CI wasn't re-read (20). 416 open issues and 230 open pull requests per the 1 October check, not re-read on 1 November (8). Seven 1.2.x patch releases since 1.2.12 with conventional commit messages, but 1.2.18 (26 October) carries a commit marked breaking (`feat(langgraph)!`, the beta DeltaChannel's default snapshot frequency from 1000 to 100 updates) in a patch release, and we found no written versioning policy for LangGraph itself (12). 1.2.19, classed 5 - Production/Stable (15).)
- Maintenance & community, LangGraph +7: BoxLang AI 83 (3.6.0 on 2026-10-20 (30). Three releases in the 90 days to 1 November, 3.4.0 on 1 September, 3.5.0 on 11 September and 3.6.0 on 20 October (20). 79 commits on the development branch in those 90 days from Luis Majano, Doug Cain, Copilot and dependency bots, with pull requests up to #326 merged on 1 November. Issue reply times couldn't be read (10 of 25). The ForgeBox package is current at 3.6.0 (15). Dependabot for Gradle monthly and Actions quarterly, tests on pull requests and development, and a format check (8 of 10). 83 in all.); LangGraph 90 (1.2.19 tagged on 2026-10-30 (30). Nine releases of the core package in the 90 days from 3 August, 1.2.11 (11 August) to 1.2.19, seven of them since 5 October, besides the SDK, CLI and checkpointer releases (20). 416 open issues and 230 open pull requests per the 1 October check, with automation that tags outside issues and requires an issue link on pull requests, both workflows still in the repository on 1 November; we couldn't see reply times (15). langgraph 1.2.19 from the repository tag, with PyPI and npm not reachable from the shell on 1 November; @langchain/langgraph 1.4.18 per the 1 October check (15). CI and CodeQL workflows present, pass state per the 1 October check (10).)