external-memory-plugin

external-memory-plugin is a skill for Claude Code, Codex from ModelEngine-Group/nexent. It costs 85 tokens per session (936 once invoked), scanned A, original, MIT.

A guide for creating plugins that connect Nexent to outside memory services. These plugins store and search information through a defined interface, with configuration, error handling, and tests.

In plain words
What is it for?
Use it when adding, changing, reviewing, or troubleshooting a memory-provider plugin. It covers the plugin manifest, provider code, network-isolated tests, deployment settings, and Mem0-based examples.
Why use it?
It helps an external memory provider behave consistently without depending on a live service during automated tests. It also reduces mistakes around tenant and user scope, credentials, timeouts, and remote failures.

Skill for Claude CodeCodex

About the project

Nexent is a zero-code platform for creating AI agents from natural-language instructions, combining tools, skills, memory, orchestration, constraints, feedback loops, and control planes. It is intended for people who need to build and deploy agents without writing orchestration code. The catalogue entries are rules and skills for working with Nexent.

ModelEngine-Group/nexent · 5,849 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/modelengine-group/nexent/external-memory-plugin
Any agent
npx skills add ModelEngine-Group/nexent --skill external-memory-plugin
Clone the repo
git clone --depth 1 https://github.com/ModelEngine-Group/nexent

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for external-memory-plugin

README.md
[![agentmods](https://agentmods.dev/badge/skills/modelengine-group/nexent/external-memory-plugin.svg)](https://agentmods.dev/skills/modelengine-group/nexent/external-memory-plugin)
Your own site
<a href="https://agentmods.dev/skills/modelengine-group/nexent/external-memory-plugin"><img src="https://agentmods.dev/badge/skills/modelengine-group/nexent/external-memory-plugin.svg" alt="Measured on agentmods" height="20"></a>
Per session 85 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 936 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00085 $0.00936
Opus 5 $0.00043 $0.00468
Sonnet 5 $0.00017 $0.00187
Haiku 4.5 $0.00009 $0.00094

Measured today against content hash 73b12333aea4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

external-memory-plugin scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/external-memory-plugin/SKILL.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

External Memory Plugin

Develop external memory adapters against the current Nexent contract and prove them without live-service CI dependencies.

Establish scope

  1. Read the repository AGENTS.md and invoke its SPEC workflow for production behavior changes.
  2. Inspect, rather than assume, these current contracts:
    • sdk/nexent/memory/providers/base.py
    • sdk/nexent/memory/models.py
    • backend/services/memory_provider_plugin_loader.py
    • backend/services/memory_external_provider_service.py
  3. Read references/plugin-contract.md before creating or reviewing a manifest or provider.
  4. Read references/mem0-example.md when implementing HTTP transport, error mapping, or tests.
  5. Preserve tenant, user, Agent, and conversation scope. Never print or persist credentials.

Implement the adapter

  1. Create <plugin-dir>/<name>/plugin.yaml and provider.py.
  2. Declare only implemented capabilities in implements.
  3. Make protocol methods asynchronous and return the exact Nexent models.
  4. Honor provider-level timeout_seconds and result limits.
  5. Map remote failures to ProviderError and the retry exception matching recovery semantics.
  6. Keep OTel attributes low-cardinality and free of queries, memory content, user identifiers, tenant identifiers, and secrets. Rely on the orchestration service's standard instrumentation unless provider-specific spans add actionable detail.
  7. Treat the deployment search/ingest switches as kill switches and the provider enabled field as the instance switch; do not bypass them in normal runtime paths.

Test without external dependencies

  1. Test manifest discovery and protocol validation with a temporary plugin directory.
  2. Mock the network at the HTTP client transport boundary; for httpx, use MockTransport.
  3. Cover successful, empty, partial, malformed, unauthorized, forbidden, rate-limited, timeout, connection, and server-error responses as applicable.
  4. Assert request scope and authentication shape without using real credentials.
  5. Assert every ingest unit receives a result and retry classification is correct.
  6. Run the narrow plugin and loader suites first, then affected service and integration suites.
  7. Reach at least 90% coverage for each new or modified module.

Read the full file on GitHub · 65 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. today First seen · 65 lines · 85 tokens per session scan A 73b12333aea4

Subscribe to this mod's changes

external-memory-plugin is a skill published in the GitHub repository ModelEngine-Group/nexent (5,849 stars, last pushed today), licensed MIT. It adds 85 tokens to every session and 936 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.