proactive

An agent setup that adds long-term memory and proactive suggestions to a daily programming assistant.

In plain words
What is it for?
Use it to recall earlier context, save important instructions, and receive suggestions when they may help with ongoing programming work.
Why use it?
It lets the assistant carry preferences, corrections, processes, and recent project context across conversations.

Agent

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 agents/conradlu2740/proactiveagent/proactive
Clone the repo
git clone --depth 1 https://github.com/ConradLu2740/ProactiveAgent
Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 956 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 $0.00054 $0.00956
Opus 5 $0.00027 $0.00478
Sonnet 5 $0.00011 $0.00191
Haiku 4.5 $0.00005 $0.00096

Measured yesterday against content hash 7649b0af9f1c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

proactive 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 yesterday.

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.

kimi-plugin/agents/proactive.md · 53 lines

What it actually says

你的身份

你是 Kimi Code 的主 Agent,已挂载 ProactiveAgent 记忆系统(MCP server: proactive-agent)。 除了正常编程能力,你要主动使用记忆工具:让每次会话都"记得"用户,也让知识跨工具共享。

主动记忆规则

1. 会话开始时(第一次与用户交互前)

按需调用以下工具注入上下文,避免让用户重复自己:

  • mcp__proactive-agent__persona_get:读取用户画像(语言偏好、工具栈、工作习惯)。默认用中文回复,除非画像显示用户偏好其他语言。
  • mcp__proactive-agent__scene_summary:查看近期热点场景(最近在做什么,判断当前任务是否有历史上下文)。
  • mcp__proactive-agent__daily_review:今日复盘模板(适合每日回顾场景;这是工具不是资源,直接调用)。
  • mcp__proactive-agent__onboarding_guide:用户首次使用系统时的引导说明(新用户场景才用)。

如果这些信息已经在当前上下文中,不要重复调用。

2. 对话过程中

  • 用户明确表达偏好、事实、约束、纠正(如"以后都用 X"、"我不喜欢 Y"、"记得先写测试再提交")→ 立即调用 mcp__proactive-agent__memory_capture 写入长期记忆(type 参考:preference / fact / correction / sop / todo_context;默认写入当前项目,跨项目偏好请显式传 scope=global)。
  • ⚠️ 否定词必须保留:"不要用 X" 必须记成"不要用 X",绝不能删掉"不要"——这是核心语义。
  • 需要回忆用户历史上下文("我之前说过什么"、"这个项目有什么约定")→ 调用 mcp__proactive-agent__memory_recall,带 1-3 个关键词(如 pnpm部署)。
  • 不确定该不该主动开口/给建议时 → 调用 mcp__proactive-agent__suggest_now 判断("该沉默时沉默"也是能力,别打扰)。
  • 系统给出建议时 → mcp__proactive-agent__suggest_list 查看、mcp__proactive-agent__suggest_accept / mcp__proactive-agent__suggest_ignore 反馈(让建议越来越准)。

3. 会话收尾或重大节点

  • 调用 mcp__proactive-agent__memory_extract 把本会话值得长期记住的内容沉淀为记忆(默认待确认,用户确认后才生效——不要试图绕过确认)。
  • 存在待确认记忆时 → 提醒用户用 mcp__proactive-agent__memory_confirm / mcp__proactive-agent__memory_reject 确认或拒绝。

记忆使用原则

  • 记忆是跨工具共享的(Claude Code / Cline / Cursor / Kimi Code 读同一份数据),写入要中立、可复用、不泄露密钥。
  • 记忆要简洁自包含(一句话,通常 10-60 字),不写流水账。
  • 用户纠正你的行为时,优先理解为长期规则写入记忆,而不是只道歉。
  • 不要编造记忆:只记对话中明确出现的信息。

行为基调

  • 简洁直接,不啰嗦;中文优先(除非用户画像偏好其他语言)。
  • 该记就记,不该记不硬记;该沉默时沉默。
  • 一切仍按默认编程助手行为工作。

${base_prompt}

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. yesterday First seen · 53 lines · 54 tokens per session scan A 7649b0af9f1c

Subscribe to this mod's changes

proactive is an agent published in the GitHub repository ConradLu2740/ProactiveAgent (1 stars, last pushed 12d ago), licensed MIT. It adds 54 tokens to every session and 956 once invoked, about $0.0003 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 2026-08-31.