engrim

The Universal Cross-Model & Cross-Agent Episodic Memory Standard.

A local-first, project-scoped SQLite memory engine that allows developers to freely switch between models and environments (Google Antigravity, Claude Code, Cursor, Codex CLI, GitHub Copilot CLI, OpenCode, and Windsurf) on the same codebase without losing architectural decisions, user constraints, or project state.

1. Overview & Core Value Proposition

"Why pay for 200,000 tokens of forgotten noise on every turn? The models are disposable utilities; your project's decisions are not."

Engrim is the universal, cross-model episodic memory standard engineered specifically for autonomous AI coding agents and heavy software development.

It solves the "200,000-token context window trap" (where coding models suffer from severe attention dilution, compounding per-turn token costs, and the loss of prior architectural decisions whenever a chat session is cleared) by replacing raw transcript replay with 4,000 characters of high-precision, curated episodic working memory stored in a single, local-first SQLite file (~/.engrim/memory.db).

The Problem vs. The Engrim Standard

Related MCP server: agentmem

2. Architecture & How It Works (The 10-Second Mental Model)

graph TD
    subgraph Agents ["Supported Agent Environments"]
        AGY["Google Antigravity<br/>(PreInvocation & Stop Hooks)"]
        CLAUDE["Claude Code<br/>(SessionStart & Stop Hooks)"]
        CURSOR["Cursor / Windsurf<br/>(Model Context Protocol stdio)"]
        CODEX["Codex CLI<br/>(Hooks, guidance & optional MCP)"]
        OPENCODE["OpenCode<br/>(Plugin & MCP)"]
        COPILOT["GitHub Copilot CLI<br/>(Command hooks & MCP)"]
    end

    subgraph CoreEngine ["engrim Core Engine (v1.4.11)"]
        ADAPTERS["Adapters & Lifecycle Hooks<br/>(agy, claude, opencode, copilot, mcp)"]
        DOCTOR["Health & Diagnostic Engine<br/>(engrim doctor --fix)"]
        PROVENANCE["Agent Provenance Engine<br/>(origin_agent tracking)"]
        ROUTER["Hybrid Retrieval & Minder<br/>(bm25 lexical + vector cosine)"]
    end

    subgraph Storage ["Local-First SQLite Store (~/.engrim/memory.db)"]
        MEMORIES[("Curated Memories<br/>(decisions, facts, feedback)")]
        FTS5["FTS5 Full-Text Search<br/>(porter stemmer, triggers)"]
        VEC["Vector Embeddings<br/>(model2vec static embeddings)"]
        LOG["Flight Recorder Log<br/>(turns + action lines)"]
    end

    AGY <-->|"hook / CLI"| ADAPTERS
    CLAUDE <-->|"hook / CLI"| ADAPTERS
    CURSOR <-->|"JSON-RPC (stdio)"| ADAPTERS
    CODEX <-->|"command hook / MCP"| ADAPTERS
    OPENCODE <-->|"plugin / MCP"| ADAPTERS
    COPILOT <-->|"command hook / MCP"| ADAPTERS
    ADAPTERS --> PROVENANCE
    PROVENANCE --> ROUTER
    ROUTER --> MEMORIES
    MEMORIES --- FTS5
    MEMORIES --- VEC
    ADAPTERS --> LOG

1. Boot: When your agent starts or a prompt is submitted, engrim injects the top-priority memory pack (~4k chars) into the agent's context, leading with [▶ RESUME HERE].

2. Capture: As the agent works, it records architectural decisions via MCP (engrim_add) or lifecycle hooks automatically, tracking the origin_agent provenance.

3. Recall & Minder: Hybrid lexical (bm25) + vector (model2vec) search retrieves relevant records on demand in ~30ms without slowing down the agent.

4. Continue-As-Clear: When context bloats or the model drifts, type /clear. The next session boots from memory.db with zero memory loss.

3. Multi-Agent Quickstart

Installation

pip install engrim

Auto-Detection (Recommended)

Run engrim setup without arguments. It automatically detects installed environments on your machine and configures them all:

engrim setup

• If ~/.gemini exists $\rightarrow$ wires Antigravity lifecycle hooks, skill, and MCP server.

• If ~/.claude exists $\rightarrow$ wires Claude Code SessionStart, Stop, status line, and CLAUDE.md.

• If ~/.cursor exists $\rightarrow$ generates and merges Cursor MCP configuration.

• If ~/.codex exists $\rightarrow$ wires Codex CLI native command hooks and managed global AGENTS.md guidance. MCP stays opt-in.

• If ~/.config/opencode exists $\rightarrow$ writes OpenCode plugin, registers MCP server, and adds AGENTS.md notes.

• If ~/.copilot exists $\rightarrow$ wires Copilot CLI hooks and the Engrim status line, registers the MCP server, and adds copilot-instructions.md notes.

Diagnostic Health Check & Self-Healing (engrim doctor)

Verify database integrity, semantic recall coverage, and all configured agent hooks across your system:

# Run comprehensive health diagnostic across all agent environments
engrim doctor

# Automatically repair broken hook paths or cross-OS migrations with self-healing PATH fallback
engrim doctor --fix

# Output diagnostic report as JSON (for CI pipelines or health monitoring)
engrim doctor --json

Real output:

================================================================================
                 🩺 ENGRIM DOCTOR: DIAGNOSTIC HEALTH CHECK                    
================================================================================
Platform   : linux (x86_64) · Python 3.12.3
Engrim CLI : /home/user/.local/bin/engrim
Project    : /workspace/my-project

[1] Database & Storage Engine
  ✓ Store Location   : /home/user/.engrim/memory.db
  ✓ Integrity Check  : ok
  ✓ Journal Mode     : wal (WAL)
  ✓ Curated Memories : 956 active / 1057 total
    (decision=413, fact=178, feedback=47, reference=20, state=283, user=15)
  ✓ Flight Log Turns : 48501 logged

[2] Semantic Recall Engine
  ✓ Model Available  : model2vec:minishlab/potion-base-8M
  ✓ Embedded Records : 1032 / 956

[3] Agent Environments & Hooks
  Google Antigravity (~/.gemini):
    ✓ PreInvocation hook: valid
    ✓ Stop hook: valid
    ✓ MCP server: /home/user/.local/bin/engrim (valid)
    ✓ Skill deployed: yes
  Claude Code (~/.claude):
    ✓ SessionStart hook: valid
    ✓ SessionEnd hook: valid
    ✓ Stop hook: valid
    ✓ UserPromptSubmit hook: valid
  Codex CLI (~/.codex):
    ✓ SessionStart hook: valid
    ✓ SessionEnd hook: valid
    ✓ Stop hook: valid
    ✓ UserPromptSubmit hook: valid
    ✓ Managed AGENTS.md guidance: present
    • MCP registration: optional, not configured
  GitHub Copilot CLI (~/.copilot):
    ✓ sessionStart hook: valid
    ✓ userPromptSubmitted hook: valid
    ✓ agentStop hook: valid
    ✓ MCP server: /home/user/.local/bin/engrim (valid)
    ✓ Status line: configured

================================================================================
Result: All systems healthy. Zero issues detected across all agent hosts.
================================================================================

• Self-Healing Path Fallback: Hook commands feature portable PATH fallback (<bin> || engrim ... || true) ensuring sessions never experience silent hook failures when directories move or dotfiles sync across machines.

• Deep Integrity Audit: Validates SQLite WAL mode, database consistency (PRAGMA integrity_check), active records, and semantic model readiness (model2vec).

Explicit Platform Setup

🪐 Google Antigravity

engrim setup --agy

• Configures ~/.gemini/config/hooks.json to execute engrim hook --agent agy --event boot on PreInvocation and engrim hook --agent agy --event stop on Stop.

• Deploys canonical Antigravity skill to ~/.gemini/config/skills/engrim/SKILL.md.

• Registers MCP server in ~/.gemini/antigravity-cli/mcp_config.json and ~/.gemini/config/mcp_config.json.

🟣 Claude Code

engrim setup --claude

• Wires SessionStart, SessionEnd, Stop, and UserPromptSubmit hooks in ~/.claude/settings.json.

• Configures live ambient status line in Claude Code's status bar.

• Appends memory usage notes to ~/.claude/CLAUDE.md.

⚡ Cursor & Windsurf

engrim setup --cursor

• Adds engrim to ~/.cursor/mcp.json running engrim serve --mcp.

For Windsurf, add engrim to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "engrim": {
      "command": "engrim",
      "args": ["serve", "--mcp"]
    }
  }
}

💻 Codex CLI

engrim setup --codex

• Wires SessionStart, SessionEnd, Stop, and UserPromptSubmit command hooks in $CODEX_HOME/hooks.json (~/.codex by default).

• Adds or updates an Engrim-owned block in $CODEX_HOME/AGENTS.md. Existing instructions and symbolic links remain unchanged. If the file already has an unmarked ## Project Memory (engrim) section, setup treats it as user-owned and does not add a duplicate. A non-empty AGENTS.override.md is not modified, but setup warns that it shadows the guidance.

• Calls the local engrim CLI directly, so MCP is not required. Hooks must be reviewed and trusted with Codex's /hooks command before they run.

To expose the engrim_* tools to Codex, enable the optional MCP server:

engrim setup --codex-mcp
# Equivalent: engrim setup --codex --codex-mcp

This uses Codex's supported codex mcp add command and preserves unrelated config.toml settings. Use /mcp to verify the server, then open a new Codex session so the hooks and guidance load.

engrim uninstall --codex removes only Engrim hooks, the managed guidance block, and the mcp_servers.engrim entry. It preserves all other Codex settings and instructions.

🤖 GitHub Copilot CLI

engrim setup --copilot

• Writes ~/.copilot/hooks/engrim.json (respects $COPILOT_HOME), a file engrim owns outright, so your other hook files are never rewritten.

  • sessionStart → the memory pack is returned as additionalContext and injected into the new session before your first prompt.

  • userPromptSubmitted → the prompt lands in the flight-recorder log. Copilot CLI discards the output of command hooks on this event, so this hook injects nothing.

  • agentStop → assistant messages are read from the supplied events.jsonl transcript path and added to the flight-recorder log.

• Registers mcpServers.engrim (engrim serve --mcp) in ~/.copilot/mcp-config.json, exposing the engrim_* tools. Records the agent writes are attributed to copilot automatically.

• Configures engrim statusline in ~/.copilot/settings.json, providing live curated, logged, in-play, and clear-safety counts. An existing custom status line is preserved.

• Appends a usage note to ~/.copilot/copilot-instructions.md.

Hook entries use Copilot's exec + args form, which runs the binary directly with no shell in between, so a path containing spaces or backslashes needs no quoting. Hook configuration is read at startup: open a new session after setup.

Copilot can flush the final assistant event shortly after agentStop begins, so Engrim polls briefly and recovers any later tail during the next session start. The event-file schema is not a public Copilot compatibility contract. Engrim detects incompatible known records without putting transcript content in diagnostics, warns once through the hook, and reports the persistent issue through engrim doctor until a compatible pass succeeds.

🟩 OpenCode

engrim setup --opencode

OpenCode has no shell hooks, so engrim ships as a plugin plus an MCP server:

• Writes ~/.config/opencode/plugins/engrim.js (respects $XDG_CONFIG_HOME). The plugin calls engrim hook --agent opencode at each lifecycle moment:

  • session boot → memory pack is injected into system prompt (once per session, and again after compaction);

  • every prompt → minder pulls the few records relevant to that message and attaches them to that user message, never to the system prompt, so the provider's prompt cache (vLLM prefix cache, Anthropic prompt caching) stays warm across turns;

  • session idle → session's new user/assistant turns land in flight-recorder log (idempotent, keyed on OpenCode message ids);

  • compaction → compaction prompt is told that durable memory lives in engrim and to list uncaptured decisions so they get engrim_add-ed.

• Registers mcp.engrim (engrim serve --mcp) in ~/.config/opencode/opencode.json, exposing engrim_* tools to the agent.

• Appends usage note to ~/.config/opencode/AGENTS.md.

OpenCode Transcript Store vs. Engrim

opencode.db stores raw transcripts (sessions, messages, parts) for OpenCode alone. Engrim provides the universal cross-session memory layer:

• Cross-Harness Continuity: Switch between OpenCode, Claude Code, Codex CLI, Cursor, or Antigravity on the same codebase from one ~/.engrim/memory.db.

• Curated Episodic State: Preserves typed records (decision, fact, state) that survive /new, session deletion, and compaction.

• Model-Driven Retrieval: Hybrid FTS5 + vector recall (engrim_recall) and per-prompt minding with cross-agent provenance (origin_agent).

🐙 GitHub Agentic Workflows (gh-aw)

See examples/gh-aw/ for engrim inside GitHub Agentic Workflows: memory across runs through artifacts and engrim merge, and a continue-as-clear restart instead of auto-compaction.

All Platforms

engrim setup --all

--all configures Codex hooks and guidance but does not opt in to the Codex MCP server. Add --codex-mcp explicitly when you want that tool interface. (Use --dry-run with any setup command to preview changes without writing to disk).

4. Empirical Proof (The 105-Session Case Study)

Tested across 105 continuous sessions on a 50,000-line algorithmic trading system. Zero regressions across 186 unit tests, zero context loss across model switches.

In production testing on an active algorithmic trading codebase running real capital:

• Over 153,000 tokens of work across days of architecture, parameter tuning, and debugging was consolidated into an active memory pack under 1,000 tokens (<1% of the context window).

• That is a 99%+ cut in reloaded context cost on every session restart.

• Seamlessly switched between Google Antigravity CLI, Claude Code, and Cursor MCP on identical repos with zero model drift or architectural regression.

5. Agent Provenance Tracking

When multiple agents collaborate on a single codebase, provenance matters. engrim records the origin of every memory entry with the origin_agent field:

• Allowed values: antigravity, claude-code, cursor, codex, opencode, copilot, cli, or user.

• Automatically populated based on the active hook, MCP client, or CLI session.

• Subtly surfaced in engrim context and engrim list:

🧠 engrim · memory restored for this project: you don't have to re-explain · /workspace
  18 of 54 curated records loaded (~3850 chars) · the rest one `recall` away

[DECISION]
- #961 [DECISION] (via Antigravity): Inverted stop loss matrix for high volatility  (risk, execution)
- #942 [DECISION] (via Claude Code): Switched primary database from MongoDB to PostgreSQL  (db, schema)
- #910 [DECISION] (via Cursor): Standardized on Pydantic v2 schemas across API boundaries  (api, types)

Existing databases are non-destructively migrated on first access via ALTER TABLE memories ADD COLUMN origin_agent TEXT.

6. Hardened Model Context Protocol (MCP) Server

Launch the zero-dependency, JSON-RPC 2.0 stdio MCP server:

engrim serve --mcp
# or: engrim mcp

stdout is strictly reserved for JSON-RPC messages, redirecting all diagnostic logs to stderr.

Core MCP Tools Exposed:

engrim_review returns safe_to_clear: null (unknown) when the project has no transcript log. With logged turns, it evaluates whether uncaptured architectural decisions exist before a developer runs /clear.

7. CLI Reference

8. Continue-As-Clear Workflow

1. Capture as you work: Whenever a major decision or architectural constraint is established, save it to memory. Connected agents do this automatically via engrim_add, or you can run engrim add.

2. Use resume-pointer: Before ending a session or clearing, add a record tagged resume-pointer describing the immediate next task. The newest pointer is pinned under [▶ RESUME HERE] at the top of the next session's boot pack. When that work is done, engrim retire marks the pointer done so finished tasks never lead later packs.

3. Verify with engrim review: Check that all recent decisions are captured before clearing.

4. Clear freely (/clear): The session window is wiped clean. engrim automatically re-injects the active memory pack on the next prompt or invocation with zero context loss.

9. Open Core Architecture: Local vs. Enterprise

Engrim maintains a strict, transparent architectural boundary between open-source single-developer productivity and enterprise infrastructure:

10. How Does Engrim Compare?

• vs gbrain: While gbrain is a provider-agnostic memory tool, engrim sets itself apart with a lightweight, local-first SQLite architecture. It requires zero cloud infrastructure, no complex daemon setup, and operates entirely on CPU.

• vs OpenCode & Codex Internal Stores: Their built-in SQLite databases store transcripts (sessions, raw message parts, lossy compaction summaries) locked to one tool. engrim is a cross-tool episodic memory engine that tracks provenance across all your tools. With the OpenCode plugin, engrim serves as the durable memory layer rather than competing with it.

• vs Pi / Personal Companions: Companion tools focus on social conversation history. engrim is engineered specifically for software engineering projects, preserving architectural decisions, invariant constraints, and technical state.

11. Security & Privacy

• 100% Local & Offline: All memory records and flight-recorder logs reside in your local SQLite file (~/.engrim/memory.db). No telemetry, no cloud sync, no tracking.

• CPU-Only Vector Inference: Uses model2vec for static embeddings (~30ms load time, no GPU required, runs on CPU). Can run pure-lexical (ENGRIM_EMBED=off) for zero extra dependencies.

• POSIX Owner Permissions: Databases are created with restricted owner-only permissions (0600).

• Git Safe: *.db is gitignored by default; memories are never accidentally committed to public version control.

12. About the Project & Author

Engrim was created by Tim Gordon (@timgordontg).

• Website: engrim.dev

• GitHub: github.com/timgordontg/engrim

• LinkedIn: linkedin.com/in/timgordon1

• Email: timgordontg@gmail.com

Tim built Engrim to establish the definitive open-source standard for agent episodic memory, giving developers complete freedom from vendor lock-in.

For enterprise licensing, advisory, pilot deployments, or custom agent integrations, reach out at timgordontg@gmail.com.

13. License

MIT © 2026 Tim Gordon.