Use cases
Developer use cases
How teams use Liminal in production — from solo dev privacy to enterprise-style workflow sweeps.
8 in-depth pages — each with comparison tables, FAQs, and install paths back to Liminal AI.
Local-first AI coding
Run an AI coding agent on your machine: session traces, tool execution, and memory on disk you control. Optional cloud LLMs or fully local inference.
Read →Refactor large codebases
Plan and execute large refactors with repo maps, grep, semantic tools, multi-file edits, tests, lint self-heal, and optional dynamic workflows across modules.
Read →Multi-file edits
Coordinated edits with edit_file, write_file, multi_file_apply, streaming large writes, and integrity checks — plus git and test follow-up.
Read →Browser automation for devs
Headless Playwright in the agent loop: navigate, snapshot, act, extract — for JS-heavy docs, QA repros, and web-backed dev tasks with stealth and session limits.
Read →Obsidian + agent memory
Connect Liminal to Obsidian: vault_write, hybrid BM25+vector recall, memory graph, workspace-scoped notes, and curator-safe pruning — a coding agent with a second brain you own.
Read →Research & intelligence workflows
Use Liminal as a research officer: web_search, parallel web_fetch, vault briefs, spawn_agent researchers, and dynamic workflows for multi-angle synthesis.
Read →Dynamic workflows
plan_workflow and run_workflow: phased sub-agents, out-of-context storage, verify gates, and distilled summaries for large parallel coding and research tasks.
Read →Multi-agent orchestration
spawn_agent, critics, contract verification, and shared context buses — coordinate specialist sub-agents without copying the parent tool registry.
Read →