superglue

tion described it like this: she gave the CLI to her agent with an instruction set and told it not to build tools, just run against the API. It handled multi-step Salesforce object creation correctly, including per-user field logic and record type templates. Concretely: instead of writing a createSalesforceOpp tool that handles contact -> account -> Opp creation with all the conditional logic, you write a skill doc and let the agent figure out which endpoints to hit and in what order. The tradeoff is: you're giving the agent more autonomy over what API calls it makes. That requires good instructions and some guardrails. But for long-tail, user-specific connectors, it's a lot more practical than building a tool for every case. Happy to discuss. Curious if others have run into the "pre-defined tool" ceiling with MCP-based connectors and how you've worked around it. Docs: https://docs.superglue.cloud/getting-started/cli-skills Repo: https://github.com/superglue-ai/superglue

missing pieces and hallucinate endpoints and parameters. 2. LLMs are stateless, but integrations are stateful:_auth handshakes, pagination, retries, multi-step flows all need memory but LLMs can’t persist intermediate values or reason across steps. 3. LLMs produce code that “looks right” but fails at runtime: LLMs cannot isolate the failing step and understand real error messages, so they can’t change what’s broken or retry with new hypotheses. 4. LLMs can’t reliably interpret imperfect API design:_humans can infer the intended function, LLMs will hallucinate what looks reasonable. We open sourced the benchmark so you can test your own APIs or contribute new ones: https://github.com/superglue-ai/superglue/tree/main/eval/llm... Curious to hear your experience, and of course always happy to share more learnings.

matically repairs integrations when upstream systems change. In short: It turns legacy integrations into code you can easily understand, test, and update. So that engineers can do more exciting feature work, and companies can migrate and upgrade systems faster. Think of it as: a context engine + code generator + integration runtime for legacy glue. What we'd love feedback on - How do you deal with "nobody knows what this script does" situations? - What would you want to know about your legacy systems? OSS/community version: https://github.com/superglue-ai/superglue More info: https://superglue.ai Happy to go deeper on the technical details.

isn’t viable in production - Our integration layer scored a 91% success rate, showing us that just throwing bigger/better LLMs at the problem won't solve it. - Only 6 out of 21 APIs worked 100% of the time, every other API had failures. - Anthropic’s models are significantly better at building API integrations than other providers. Here is the results chart: https://superglue.ai/files/performance.png What made LLMs fail: - Lack of context (LLMs are just not great at understanding what API endpoints exist and what they do, even if you give them documentation which we did) - Multi-step workflows (chaining API calls) - Complex API design: APIs like Square, PostHog, Asana (Forcing project selection among other things trips llms over) We've open-sourced the benchmark so you can test any API and see where it ranks: https://github.com/superglue-ai/superglue/tree/main/packages... Check out the repo, consider giving it a star, or see the full ranking at https://superglue.ai/api-ranking/. If you're building agents that need reliable API access, we'd love to hear your approach, or you can try our integration layer at superglue.ai. Next up: benchmarking MCP.

exist and what they do, even if you give them documentation which we did) - Multi-step workflows (chaining API calls) - Complex API design: APIs like Square, PostHog, Asana (Forcing project selection among other things trips llms over) We've open-sourced the benchmark so you can test any API and see where it ranks: https://github.com/superglue-ai/superglue/tree/main/packages... Check out the repo, consider giving it a star, or see the full ranking at https://superglue.ai/api-ranking/ If you're building agents that need reliable API access, we'd love to hear your approach - or you can try our integration layer at superglue.ai. Next up: benchmarking MCP.

bly, like fetching transactions in Stripe and updating them in Hubspot, to build agents that react to app triggers or to wrap your own API into task-specific tools agents can use. Quickstart guide: https://docs.superglue.cloud/mcp Find us on GitHub: https://github.com/superglue-ai/superglue Would be curious to hear how you’ve been dealing with tool use, and what you would want to build with MCPglue.

let your agent run cross-API workflows reliably, like fetching transactions in Stripe and updating them in Hubspot, to build ambient agents that react to app triggers or to wrap your own API into task-specific tools agents can use. Repo: https://github.com/superglue-ai/superglue Quickstart guide: https://docs.superglue.cloud/mcp We've seen agents handle max 10 tools at once. Would be curious to hear how many tools your agents can handle on average. And ofc what you would want to build with MCPglue.

PI configuration by analyzing API docs. - Handles pagination, authentication, and error retries. - Transforms response data into the exact schema you want using JSONata expressions. - Validates that all data coming through follows that schema, and fixes transformations when they break. We built this after noticing how much of our team's time was spent building and maintaining data integration code. Our approach is a bit different to other solutions out there because we (1) use LLMs to generate mapping code, so you can basically build your own universal API with the exact fields that you need, and (2) validate that what you get is what you’re supposed to get, with the ability to “self-heal” if anything goes wrong. You can run superglue yourself (https://github.com/superglue-ai/superglue - license is GPL), or you can use our hosted version (https://app.superglue.cloud) and our TS SDK (npm i @superglue/client). Here’s a quick demo: https://www.youtube.com/watch?v=A1gv6P-fas4 You can also try out Jira and Shopify demos on our website (https://superglue.cloud) Excited to share superglue with everyone here—it's early so you'll probably find bugs, but we'd love to get your thoughts and see if others find this approach useful!