Magic Loops

What Magic Loops does

Magic Loops turns a plain-language request into an App or Loop made of executable blocks. Its practical difference from prompt-only builders is the mix of LLM reasoning and generated code: classification or drafting can use a model, while parsing, arithmetic, API requests, and branching can run deterministically. A loop can run manually, at a URL, from a webhook, or on a schedule.

The vendor says every paid plan costs $1 USD per 1,000 credits. It lists Hobby at $9 for 9,000 credits, Professional at $20 for 20,000, Developer at $50 for 50,000, and Startup at $300 for 300,000. Free provides 250 generation credits daily and 500 run credits monthly, but free loops are public. Published examples estimate a username generator at 4–5 credits per run, an SEC filing notifier at 50–100, and a scraping-plus-message workflow at 75–144. Test at least 100 runs and track successful completions, median and p95 credits, retries, latency, and failures by block.

Pricing, alternatives, and fit

Published pricing is: Free is $0; Hobby $9/month for 9,000 credits; Professional $20 for 20,000; Developer $50 for 50,000; Startup $300 for 300,000. Pricing should be evaluated as total cost per accepted result, not only the headline fee. Include model calls, external APIs, compute, storage, retries, human review, implementation, and support. Run a capped pilot with representative peak volume and record successful completions, p50 and p95 latency, correction minutes, failures by cause, and spend per approved output.

Relevant comparisons are Gumloop, n8n, Zapier, BuildShip. Compare them on the same inputs because category labels hide differences in hosting, permissions, integrations, observability, and exportability. The best choice is the one that meets a written quality and recovery threshold on the buyer's own workload, not the one with the longest feature list.

Pros, cons, and implementation guidance

Concrete strengths include Transparent $1 per 1,000 credit pricing; Generated code handles deterministic logic; Fast path to callable or scheduled automation; Credit examples aid cost estimates. Important limitations include Free loops are public; Credit use varies by block and service; Product is labeled beta; Generated-code debugging can exceed no-code expectations. These tradeoffs need owners. Security review should cover SSO, role-based access, encryption, secrets, audit logs, subprocessors, retention, deletion, incident response, and whether customer data trains shared models. Technical review should test rate limits, timeouts, retries, idempotency, concurrency, versioning, partial failure, export, and recovery after interruption.

Practical use cases are Scheduled API summaries; Classification and notification routing; Internal-tool prototypes; Deterministic parsing plus LLM extraction. Start with one bounded workflow whose current time, cost, and error rate are known. Use at least 50 real examples plus malformed and adversarial inputs. Keep a manual fallback until the system repeatedly meets the acceptance bar. For tools that execute actions, use least privilege and require human confirmation for money movement, publication, deletion, account changes, and external messages. For generated media or extracted documents, preserve provenance and source references so reviewers can verify consequential output.