ReMe Episodic Memory & Causal Reflection Engine

@agentscope-ai/reme · v1.2.0

A causal episodic memory and self-reflection engine for DeepSeek Harness. Models chronological event streams and causal chains to isolate past failures and prevent repeated mistakes.

Episodic-MemoryCausal-GraphReflectionSelf-Correction
GitHub Stars
31 K+
+12.4% this month
Monthly Downloads
119 K+
Monthly registry pulls
Reach Score
98.2/ 100
Top Tier Ecosystem
Runtime
Cordis v3+
Node 18+ / Bun / Deno

Installation & Integration

CLI one-click launch, package managers, and Cordis integration

bash
manager:
$ pnpm add @agentscope-ai/reme

Architecture

ReMe Memory endows autonomous agents with genuine retrospective wisdom. When navigating long-horizon software engineering tasks, agents often forget early failed attempts, falling into repetitive loops and retrying already-disproven solutions. This plugin implements a causal episodic event graph inside Cordis. Every significant decision, failed assertion, and subsequent breakthrough serializes as an episodic node connected by causal edges. When the agent navigates similar decision forks, ReMe resurfaces relevant historical post-mortems, ensuring the agent learns continuously from past mistakes.

Architectural Principles & Constraints

01
Strict Type Isolation

Guaranteed by TypeScript compile-time contracts, inter-plugin event bus calls enjoy zero-drift safety.

02
Sub-Millisecond Hot Reload

Supports dynamic runtime mounting and graceful unloading without restarting the primary host process.

03
Deterministic State Machine

Embeds multi-phase execution lifecycle guards, preventing context loss during long-horizon reasoning.

04
Zero Native Build Dependencies

Designed for lightweight cross-platform environments, booting instantly across Node.js, Bun, and Deno.

Core Features

01
Causal Graph Topology: Stores verified cause-and-effect loops rather than detached text
02
Zero-Repetition Guard: Flags previously attempted dead-ends, preventing repeated blunders
03
Chronological Post-Mortem Playback: Step through the agent's historical mental deductions
04
Compact Episodic Indexing: Scalable graph structure indexing tens of thousands of past turns

Core Workflow

01

Causal Event Capture

Logs agent actions alongside their initial hypotheses and actual sandbox outcomes.

02

Causal Graph Modeling

Links failures to their effective resolutions, compiling episodic cause-effect nodes.

03

Dilemma-Driven Retrieval

Retrieves past post-mortems when current trajectories stall or repeat errors.

04

Negative Constraint Injection

Injects verified past lessons as explicit negative constraints into active prompts.

Configuration Parameters Reference (YAML / JSON)

ParameterTypeDefaultDescription
maxEpisodicNodesnumber1000Maximum episodic event nodes retained
causalSimilarityThresholdnumber0.82Similarity threshold to trigger post-mortem recall
autoInjectNegativeConstraintsbooleantrueAuto-inject past failures as negative prompts

Use Cases

Enterprise Production Agent

Relies on microkernel lifecycle guards and fault-tolerant state machines for continuous reliability.

SWE-bench Benchmark Evaluation

Native integration with SWE-bench workflows, automatically capturing diffs and verification metrics.

Autonomous Code Refactoring

Separates reasoning from tool actions to independently locate and refactor multi-file codebases.

Cross-Tool Workflow Automation

Safely orchestrates events across sandboxes to seamlessly link enterprise developer tooling.

Best Practices

01
Sandbox Permission Guard

Strictly isolate sub-process calls and network scope; deploy within Docker containers in production.

02
Exponential Backoff Retries

Configure adaptive exponential retries with strict timeouts to mitigate upstream model rate limits.

03
Session State Checkpointing

Persist state machine snapshots to survive hardware interruptions and resume instantly without loss.

04
Full Trajectory Audit Logs

Enable full trace logging, aggregating reasoning thought streams and tool I/O into your observability hub.

FAQ

Q1:How to handle timeouts in long-running autonomous tasks?

Increase the timeout parameter inside your YAML configuration and dispatch periodic heartbeat signals across the Cordis event bus. For long-running tool execution and model reasoning, configure persistent session snapshotting so suspended tasks can safely resume their exact context after interruptions, preventing the kernel from recycling active agent sessions prematurely.

Q2:How to capture and stream the model reasoning thought process?

The harness runtime natively provides end-to-end streaming hooks while its state machine automatically intercepts and strips <think> reasoning tags from model responses. Subscribe directly to the onThink event listener to consume live reasoning token streams in real-time, delivering typewriter animation to the user interface while persisting full trajectories for audit compliance.

Q3:How to resolve dependency conflicts across multiple plugins?

Cordis microkernel uses directed acyclic graph topological sorting to dynamically resolve plugin dependencies. When shared services or runtime versions conflict, assign distinct isolated namespaces at the application entrypoint. Leveraging context injection alongside lazy activation ensures dependencies load strictly on-demand when tools are triggered, maintaining system stability and preventing memory bloat.

Q4:How to enforce permissions and sandbox isolation in production?

Combine isolated container sandboxing with fine-grained capability checks to prevent plugins from accessing sensitive files or unauthorized external networks. Enforce explicit runtime system call whitelisting through the microkernel, executing all third-party tool scripts inside isolated ephemeral containers to block arbitrary code execution and eliminate security privilege escalation risks entirely.

Recommended Ecosystem Plugins

Explore related Cordis extensions designed to work synergistically