# Slowave

- [Slowave: Cross-Session Memory for AI Coding Agents](https://slowave.ai/index.md): Slowave is a local, brain-inspired memory layer that gives AI coding agents durable, cross-tool context — no LLM API key required, fully inspectable.
- [Quickstart: Install and Configure Slowave in 5 Minutes](https://slowave.ai/quickstart.md): Install Slowave, configure your AI coding clients, verify the setup, and watch your agent build durable project memory from its very first task.
- [How Slowave Memory Works: The Five-Verb MCP Lifecycle](https://slowave.ai/how-it-works.md): A deep-dive into Slowave's five-verb MCP lifecycle, local memory types, background consolidation, and how salience and feedback evolve memory over time.
- [Install Slowave with pipx or Homebrew, then run setup](https://slowave.ai/setup/installation.md): Install Slowave via pipx or Homebrew, preview changes with --dry-run, run slowave setup to wire every detected client, and verify with slowave doctor.
- [Configure each AI coding client to connect to Slowave](https://slowave.ai/setup/client-integrations.md): Per-client setup details for Claude Code, Cline, Cursor, Windsurf, OpenCode, Codex, and Claude Desktop — config files, detection, and manual steps.
- [Slowave background services: daemon, worker, and backup](https://slowave.ai/setup/services.md): Three services power Slowave at runtime — an HTTP MCP daemon, a consolidation worker, and a daily backup. Learn what each does and how to verify them.
- [Uninstall, purge, or remove the Slowave package fully](https://slowave.ai/setup/removal.md): Remove Slowave integrations while keeping your memories, wipe all local data, or uninstall the package entirely — with dry-run support at every step.
- [From agent sessions to durable memory: the full lifecycle](https://slowave.ai/guides/memory-lifecycle.md): How raw session events become episodes, prototypes, and durable schemas — and how feedback, salience, and decay keep memory honest over time.
- [Scoping memory to projects, domains, and workflows](https://slowave.ai/guides/scopes.md): How scopes isolate memory by context, when generalized schemas cross scope boundaries, and best practices for naming and managing your scope strategy.
- [Capture and reuse multi-step procedures with Slowave](https://slowave.ai/guides/procedures.md): How execution-backed procedures are captured at commit, surfaced during activation, and refined through use and effect feedback over repeated sessions.
- [Inspect and manage memory with the local dashboard](https://slowave.ai/guides/dashboard.md): Browse the local web dashboard to inspect memories, schemas, procedures, retrievals, and activity — use Forget and Unforget to manage what agents recall.
- [Slowave CLI reference: commands, subcommands, and flags](https://slowave.ai/operations/cli-reference.md): Complete reference for every Slowave CLI command — lifecycle, session, operations, maintenance, setup, and diagnostics — with real invocation examples.
- [Slowave database backup, restore, and migration guide](https://slowave.ai/operations/backup-restore.md): How to create, schedule, and restore gzip-compressed SQLite snapshots of your Slowave memory database, including legacy migration from ~/.slowave.
- [Slowave configuration and environment variable reference](https://slowave.ai/operations/configuration.md): All environment variables that control Slowave's runtime data location, HTTP daemon, backup retention, session timeouts, and native library compatibility.
- [Troubleshoot common Slowave installation and runtime issues](https://slowave.ai/operations/troubleshooting.md): Solutions for Slowave daemon failures, worker stalls, dashboard problems, client integration, and database errors — always start with slowave doctor.
- [Slowave architecture: memory model and cognitive cycle](https://slowave.ai/reference/architecture.md): How Slowave stores, consolidates, and retrieves memory - covering the five-verb MCP lifecycle, local components, scopes, lifecycle states, and procedures.
- [Slowave design: rationale, boundaries, and positioning](https://slowave.ai/reference/design.md): Why Slowave separates memory from reasoning, what it remembers, where its boundaries lie, and how language models interact with the memory layer.
- [Slowave benchmark results and evaluation methodology](https://slowave.ai/reference/benchmarks.md): LoCoMo and LongMemEval oracle results for Slowave's LLM-free retrieval layer, with methodology, category breakdowns, and reproduction commands.
- [Slowave MCP Tools: Five-Verb Cognitive Cycle Guide](https://slowave.ai/mcp/overview.md): A guide to Slowave's five MCP tools — activate, remember, recall, feedback, and commit — their ordering, server config, and session lifecycle.
- [slowave_activate: Prime Working Memory at Task Start](https://slowave.ai/mcp/activate.md): Call slowave_activate at the start of every task to open a session, run spreading activation, and surface relevant memories and procedures for the scope.
- [slowave_remember: store durable typed claims in memory](https://slowave.ai/mcp/remember.md): Use slowave_remember to persist typed facts, decisions, constraints, and lessons into long-term memory so they surface in future sessions and activations.
- [slowave_recall: Semantic Mid-Task Memory Retrieval](https://slowave.ai/mcp/recall.md): Use slowave_recall for deliberate mid-task lookups when the task evolves and activation alone did not surface the historical context you need.
- [slowave_feedback: record memory and procedure assessments](https://slowave.ai/mcp/feedback.md): Submit append-only assessments of retrieved memories and procedures — required for every exposed retrieval target before slowave_commit will succeed.
- [slowave_commit: Close the Task and Persist Outcomes](https://slowave.ai/mcp/commit.md): Call slowave_commit at task end to record the outcome and verification, optionally capture a reusable procedure, and trigger offline memory consolidation.