feat(setup): add CodeBuddy integration

README.md

@@ -133,6 +133,16 @@ or `~/.qoder/`. QoderWork uses its native user config at `~/.qoderwork/`.
 Both integrations register `SessionStart`, `UserPromptSubmit`, and `Stop`
 hooks in `settings.json`.
 
+### [CodeBuddy](https://www.codebuddy.cn/)
+
+```bash
+mnemon setup --target codebuddy --yes
+```
+
+CodeBuddy deploys the mnemon skill, prompt files, and native hooks to
+`.codebuddy/` or `~/.codebuddy/`. The integration registers `SessionStart`,
+`UserPromptSubmit`, and `Stop` hooks in `settings.json`.
+
 ### [OpenClaw](https://github.com/openclaw/openclaw)
 
 ```bash
@@ -233,7 +243,7 @@ memory is useful.
 
 - **Zero user-side operation** — install once; supported runtimes can use hooks, minimal runtimes can use persistent rules
 - **LLM-supervised** — the host LLM decides what to remember, update, and forget; no embedded LLM, no API keys
-- **Multi-framework support** — Claude Code, Codex, Cursor, TRAE/TRAE Work, Qoder/QoderWork, and Hermes Agent (hooks), OpenClaw (plugins), Pi (extensions), Nanobot (skills), and more
+- **Multi-framework support** — Claude Code, Codex, Cursor, TRAE/TRAE Work, Qoder/QoderWork, CodeBuddy, and Hermes Agent (hooks), OpenClaw (plugins), Pi (extensions), Nanobot (skills), and more
 - **Markdown-installable harness** — `SKILL.md`, `INSTALL.md`, `GUIDELINE.md`, and four lifecycle reminders
 - **Four-graph architecture** — temporal, entity, causal, and semantic edges, not just vector similarity
 - **Intent-native protocol** — three primitives (`remember`, `link`, `recall`) map to the LLM's cognitive vocabulary, not database syntax; structured JSON output with signal transparency
@@ -262,6 +272,8 @@ All your local agentic AIs — across sessions and frameworks — sharing one po
                 │
   QoderWork ────┤
                 │
+  CodeBuddy ────┤
+                │
   Hermes Agent ─┤
                 │
   OpenClaw ─────┤
@@ -279,11 +291,11 @@ All your local agentic AIs — across sessions and frameworks — sharing one po
 
 The foundation is in place: a single `~/.mnemon` database that any agent can
 read and write. Claude Code, Codex, Cursor, TRAE/TRAE Work, Qoder/QoderWork,
-and Hermes Agent setup automate hook installation; OpenClaw can use plugin
-hooks; Pi integrates via native skills and TypeScript lifecycle extensions;
-Nanobot integrates via skill files; NanoClaw integrates via container skills and
-volume mounts. The same harness can be installed in any LLM CLI that supports
-skills, rules, system prompts, or event hooks.
+CodeBuddy, and Hermes Agent setup automate hook installation; OpenClaw can use
+plugin hooks; Pi integrates via native skills and TypeScript lifecycle
+extensions; Nanobot integrates via skill files; NanoClaw integrates via
+container skills and volume mounts. The same harness can be installed in any LLM
+CLI that supports skills, rules, system prompts, or event hooks.
 
 The longer-term direction is a **memory gateway**: protocol decoupled from storage engine. The current SQLite backend is the first adapter; the protocol surface (`remember / link / recall`) can sit on top of PostgreSQL, Neo4j, or any graph database. Agent-side optimization (when to recall, what to remember) and storage-side optimization (indexing, graph algorithms) evolve independently. See [Future Direction](docs/design/08-decisions.md#82-future-direction) for details.
 

cmd/setup.go

@@ -22,11 +22,11 @@ var setupCmd = &cobra.Command{
 	Short: "Deploy mnemon into LLM CLI environments",
 	Long: `Detect installed LLM CLIs and deploy mnemon integration.
 
-By default, installs to project-local config (.claude/, .codex/, .cursor/, .trae/, .qoder/, .openclaw/, .nanobot/, .pi/).
-Use --global to install to user-wide config (~/.claude/, ~/.codex/, ~/.cursor/, ~/.trae/, ~/.qoder/, ~/.openclaw/, ~/.nanobot/workspace/, ~/.pi/agent/).
+By default, installs to project-local config (.claude/, .codex/, .cursor/, .trae/, .qoder/, .codebuddy/, .openclaw/, .nanobot/, .pi/).
+Use --global to install to user-wide config (~/.claude/, ~/.codex/, ~/.cursor/, ~/.trae/, ~/.qoder/, ~/.codebuddy/, ~/.openclaw/, ~/.nanobot/workspace/, ~/.pi/agent/).
 Hermes Agent and QoderWork use their native user config at ~/.hermes/ and ~/.qoderwork/.
 
-Supported environments: Claude Code, Codex, Cursor, Trae, Qoder, QoderWork, OpenClaw, Nanobot, Pi, Hermes Agent.
+Supported environments: Claude Code, Codex, Cursor, Trae, Qoder, QoderWork, CodeBuddy, OpenClaw, Nanobot, Pi, Hermes Agent.
 
 Examples:
   mnemon setup                              # Interactive: project-local install
@@ -36,6 +36,7 @@ Examples:
   mnemon setup --target trae                # Non-interactive: Trae skill and hooks
   mnemon setup --target qoder               # Non-interactive: Qoder skill and hooks
   mnemon setup --target qoderwork           # Non-interactive: QoderWork skill and hooks
+  mnemon setup --target codebuddy           # Non-interactive: CodeBuddy skill and hooks
   mnemon setup --target hermes              # Non-interactive: Hermes Agent only
   mnemon setup --eject                      # Interactive: remove integrations
   mnemon setup --eject --target claude-code # Non-interactive: remove Claude Code only
@@ -44,16 +45,16 @@ Examples:
 }
 
 func init() {
-	setupCmd.Flags().StringVar(&setupTarget, "target", "", "target environment (claude-code, codex, cursor, trae, qoder, qoderwork, openclaw, nanobot, pi, hermes)")
+	setupCmd.Flags().StringVar(&setupTarget, "target", "", "target environment (claude-code, codex, cursor, trae, qoder, qoderwork, codebuddy, openclaw, nanobot, pi, hermes)")
 	setupCmd.Flags().BoolVar(&setupEject, "eject", false, "remove mnemon integrations")
 	setupCmd.Flags().BoolVar(&setupYes, "yes", false, "auto-confirm all prompts (CI-friendly)")
 	setupCmd.Flags().BoolVar(&setupGlobal, "global", false, "install to user-wide config instead of project-local config")
 	rootCmd.AddCommand(setupCmd)
 }
 
 func runSetup(cmd *cobra.Command, args []string) error {
-	if setupTarget != "" && setupTarget != "claude-code" && setupTarget != "codex" && setupTarget != "cursor" && setupTarget != "trae" && setupTarget != "qoder" && setupTarget != "qoderwork" && setupTarget != "openclaw" && setupTarget != "nanobot" && setupTarget != "pi" && setupTarget != "hermes" {
-		return fmt.Errorf("invalid target %q (must be claude-code, codex, cursor, trae, qoder, qoderwork, openclaw, nanobot, pi, or hermes)", setupTarget)
+	if setupTarget != "" && setupTarget != "claude-code" && setupTarget != "codex" && setupTarget != "cursor" && setupTarget != "trae" && setupTarget != "qoder" && setupTarget != "qoderwork" && setupTarget != "codebuddy" && setupTarget != "openclaw" && setupTarget != "nanobot" && setupTarget != "pi" && setupTarget != "hermes" {
+		return fmt.Errorf("invalid target %q (must be claude-code, codex, cursor, trae, qoder, qoderwork, codebuddy, openclaw, nanobot, pi, or hermes)", setupTarget)
 	}
 
 	envs := setup.DetectEnvironments(setupGlobal)
@@ -89,7 +90,7 @@ func runInstallFlow(envs []setup.Environment) error {
 
 	if len(detected) == 0 {
 		fmt.Println("\nNo supported LLM CLI environments detected.")
-		fmt.Println("Install Claude Code, Codex, Cursor, Trae, Qoder, QoderWork, OpenClaw, Nanobot, Pi, or Hermes Agent, then run 'mnemon setup' again.")
+		fmt.Println("Install Claude Code, Codex, Cursor, Trae, Qoder, QoderWork, CodeBuddy, OpenClaw, Nanobot, Pi, or Hermes Agent, then run 'mnemon setup' again.")
 		return nil
 	}
 
@@ -141,6 +142,8 @@ func installEnv(env *setup.Environment) error {
 		err = installQoder(env)
 	case "qoderwork":
 		err = installQoderWork(env)
+	case "codebuddy":
+		err = installCodeBuddy(env)
 	case "openclaw":
 		err = installOpenClaw(env)
 	case "nanobot":
@@ -678,6 +681,83 @@ func installQoderLike(configDir string, writeSkill func(string) (string, error),
 	return nil
 }
 
+// ─── CodeBuddy ──────────────────────────────────────────────────────
+
+func installCodeBuddy(env *setup.Environment) error {
+	configDir := env.ConfigDir
+
+	if !setupGlobal && !setupYes && setup.IsInteractive() {
+		home := setup.HomeDir()
+		localDir := ".codebuddy"
+		globalDir := home + "/.codebuddy"
+		idx := setup.SelectOne("Install scope",
+			[]string{
+				fmt.Sprintf("Local — this project only (%s/)", localDir),
+				fmt.Sprintf("Global — all projects (%s/)", globalDir),
+			}, 0)
+		if idx == 1 {
+			configDir = globalDir
+		} else {
+			configDir = localDir
+		}
+	}
+
+	fmt.Printf("\nSetting up CodeBuddy (%s)...\n", configDir)
+
+	fmt.Println("\n[1/3] Skill")
+	if path, err := setup.CodeBuddyWriteSkill(configDir); err != nil {
+		setup.StatusError(0, 0, "Skill", err)
+		return err
+	} else {
+		setup.StatusOK(0, 0, "Skill", path)
+	}
+
+	fmt.Println("\n[2/3] Prompts")
+	var promptPath string
+	if path, err := setup.WritePromptFiles(); err != nil {
+		setup.StatusError(0, 0, "Prompts", err)
+		return err
+	} else {
+		setup.StatusOK(0, 0, "Prompts", path)
+		promptPath = path
+	}
+
+	fmt.Println("\n[3/3] Hooks")
+	for _, hook := range []struct {
+		label    string
+		filename string
+		content  []byte
+	}{
+		{"Hook: prime", "prime.sh", assets.CodeBuddyPrimeHook},
+		{"Hook: remind", "user_prompt.sh", assets.CodeBuddyUserPromptHook},
+		{"Hook: nudge", "stop.sh", assets.CodeBuddyStopHook},
+	} {
+		if path, err := setup.CodeBuddyWriteHook(configDir, hook.filename, hook.content); err != nil {
+			setup.StatusError(0, 0, hook.label, err)
+			return err
+		} else {
+			setup.StatusOK(0, 0, hook.label, path)
+		}
+	}
+	if path, err := setup.CodeBuddyRegisterHooks(configDir); err != nil {
+		setup.StatusError(0, 0, "Settings", err)
+		return err
+	} else {
+		setup.StatusUpdated(0, 0, "Settings", path)
+	}
+
+	fmt.Println()
+	fmt.Println("Setup complete!")
+	fmt.Printf("  Skill    %s/skills/mnemon/SKILL.md\n", configDir)
+	fmt.Printf("  Hooks    %s/settings.json (SessionStart, UserPromptSubmit, Stop)\n", configDir)
+	fmt.Printf("  Prompts  %s/ (guide.md, skill.md)\n", promptPath)
+	fmt.Println()
+	fmt.Println("Restart CodeBuddy Code to activate the mnemon skill and hooks.")
+	fmt.Println("Run 'mnemon setup --eject --target codebuddy' to remove.")
+
+	return nil
+}
+
 // ─── OpenClaw ───────────────────────────────────────────────────────
 
 func installOpenClaw(env *setup.Environment) error {
@@ -1133,6 +1213,13 @@ func ejectEnv(env *setup.Environment) error {
 			return errs[0]
 		}
 
+	case "codebuddy":
+		errs := setup.CodeBuddyEject(env.ConfigDir)
+		ejectMarkdown("AGENTS.md", "Remove memory guidance from ./AGENTS.md?")
+		if len(errs) > 0 {
+			return errs[0]
+		}
+
 	case "openclaw":
 		errs := setup.OpenClawEject(env.ConfigDir)
 		ejectMarkdown("AGENTS.md", "Remove memory guidance from ./AGENTS.md?")

docs/USAGE.md

@@ -34,6 +34,7 @@ mnemon setup --target cursor
 mnemon setup --target trae
 mnemon setup --target qoder
 mnemon setup --target qoderwork
+mnemon setup --target codebuddy
 mnemon setup --target openclaw
 mnemon setup --target pi
 mnemon setup --target nanobot --global
@@ -50,7 +51,7 @@ mnemon setup --eject --target claude-code
 | Flag | Default | Description |
 |---|---|---|
 | `--global` | `false` | Install to user-wide config instead of project-local (recommended for Nanobot: installs to `~/.nanobot/workspace/`; Pi installs to `~/.pi/agent/`; Hermes installs to `~/.hermes/`; QoderWork installs to `~/.qoderwork/`) |
-| `--target <name>` | (auto-detect) | Target environment: `claude-code`, `codex`, `cursor`, `trae`, `qoder`, `qoderwork`, `openclaw`, `nanobot`, `pi`, or `hermes` |
+| `--target <name>` | (auto-detect) | Target environment: `claude-code`, `codex`, `cursor`, `trae`, `qoder`, `qoderwork`, `codebuddy`, `openclaw`, `nanobot`, `pi`, or `hermes` |
 | `--eject` | `false` | Remove mnemon integrations |
 | `--yes` | `false` | Auto-confirm all prompts |
 

docs/zh/README.md

@@ -113,6 +113,16 @@ Qoder 会将 mnemon skill、prompt 文件和原生 hooks 部署到 `.qoder/` 或
 `~/.qoder/`。QoderWork 使用原生用户级配置 `~/.qoderwork/`。两者都会在
 `settings.json` 中注册 `SessionStart`、`UserPromptSubmit` 和 `Stop` hooks。
 
+### [CodeBuddy](https://www.codebuddy.cn/)
+
+```bash
+mnemon setup --target codebuddy --yes
+```
+
+CodeBuddy 会将 mnemon skill、prompt 文件和原生 hooks 部署到 `.codebuddy/`
+或 `~/.codebuddy/`。该集成会在 `settings.json` 中注册 `SessionStart`、
+`UserPromptSubmit` 和 `Stop` hooks。
+
 ### [OpenClaw](https://github.com/openclaw/openclaw)
 
 ```bash
@@ -196,7 +206,7 @@ Agent 工作，并且只在有用时调用 Mnemon
 
 - **零用户操作** — 安装一次；支持 hook 的 runtime 可用 hook，minimal runtime 可用持久规则
 - **LLM 监督式** — 宿主 LLM 主动决定记什么、更新什么、遗忘什么；无内嵌 LLM，无 API 密钥
-- **多框架支持** — Claude Code、Codex、Cursor、TRAE/TRAE Work、Qoder/QoderWork 和 Hermes Agent（hooks）、OpenClaw（plugins）、Pi（extensions）、Nanobot（skills）等
+- **多框架支持** — Claude Code、Codex、Cursor、TRAE/TRAE Work、Qoder/QoderWork、CodeBuddy 和 Hermes Agent（hooks）、OpenClaw（plugins）、Pi（extensions）、Nanobot（skills）等
 - **Markdown 可安装 harness** — `SKILL.md`、`INSTALL.md`、`GUIDELINE.md` 和四个生命周期提醒
 - **四图架构** — 时序、实体、因果、语义四种边，不仅仅是向量相似度
 - **意图原生协议** — 三个原语（`remember`、`link`、`recall`）映射到 LLM 的认知词汇而非数据库语法；结构化 JSON 输出，带信号透明度
@@ -224,6 +234,8 @@ Agent 工作，并且只在有用时调用 Mnemon
                 │
   QoderWork ────┤
                 │
+  CodeBuddy ────┤
+                │
   Hermes Agent ─┤
                 │
   OpenClaw ─────┤
@@ -239,7 +251,7 @@ Agent 工作，并且只在有用时调用 Mnemon
   Gemini CLI ───┘
 ```
 
-基础已就绪：一个 `~/.mnemon` 数据库，任何 agent 都可以读写。Claude Code、Codex、Cursor、TRAE/TRAE Work、Qoder/QoderWork 和 Hermes Agent setup 可自动安装 hook；OpenClaw 可以使用 plugin hooks；Pi 通过原生 skill 和 TypeScript lifecycle extension 集成；Nanobot 通过 skill 文件集成；NanoClaw 通过容器技能和卷挂载集成。同一个 harness 可以安装到任何支持 skill、rule、system prompt 或 event hook 的 LLM CLI。
+基础已就绪：一个 `~/.mnemon` 数据库，任何 agent 都可以读写。Claude Code、Codex、Cursor、TRAE/TRAE Work、Qoder/QoderWork、CodeBuddy 和 Hermes Agent setup 可自动安装 hook；OpenClaw 可以使用 plugin hooks；Pi 通过原生 skill 和 TypeScript lifecycle extension 集成；Nanobot 通过 skill 文件集成；NanoClaw 通过容器技能和卷挂载集成。同一个 harness 可以安装到任何支持 skill、rule、system prompt 或 event hook 的 LLM CLI。
 
 更长远的方向是**记忆网关**：协议层与存储引擎解耦。当前 SQLite 后端是第一个适配器；协议面（`remember / link / recall`）可运行在 PostgreSQL、Neo4j 或任何图数据库之上。Agent 侧优化（何时召回、记什么）与存储侧优化（索引、图算法）独立演进。详见[未来方向](design/08-decisions.md#82-未来方向)。
 

docs/zh/USAGE.md

@@ -34,6 +34,7 @@ mnemon setup --target cursor
 mnemon setup --target trae
 mnemon setup --target qoder
 mnemon setup --target qoderwork
+mnemon setup --target codebuddy
 mnemon setup --target openclaw
 mnemon setup --target pi
 mnemon setup --target nanobot --global
@@ -50,7 +51,7 @@ mnemon setup --eject --target claude-code
 | 标志 | 默认值 | 说明 |
 |---|---|---|
 | `--global` | `false` | 安装到用户级配置而非项目本地（Nanobot 推荐安装到 `~/.nanobot/workspace/`；Pi 安装到 `~/.pi/agent/`；Hermes 安装到 `~/.hermes/`；QoderWork 安装到 `~/.qoderwork/`） |
-| `--target <name>` | (自动检测) | 目标环境：`claude-code`、`codex`、`cursor`、`trae`、`qoder`、`qoderwork`、`openclaw`、`nanobot`、`pi` 或 `hermes` |
+| `--target <name>` | (自动检测) | 目标环境：`claude-code`、`codex`、`cursor`、`trae`、`qoder`、`qoderwork`、`codebuddy`、`openclaw`、`nanobot`、`pi` 或 `hermes` |
 | `--eject` | `false` | 移除 mnemon 集成 |
 | `--yes` | `false` | 自动确认所有提示 |
 

internal/setup/assets/assets.go

@@ -71,6 +71,18 @@ var QoderUserPromptHook []byte
 //go:embed qoder/stop.sh
 var QoderStopHook []byte
 
+//go:embed codebuddy/SKILL.md
+var CodeBuddySkill []byte
+
+//go:embed codebuddy/prime.sh
+var CodeBuddyPrimeHook []byte
+
+//go:embed codebuddy/user_prompt.sh
+var CodeBuddyUserPromptHook []byte
+
+//go:embed codebuddy/stop.sh
+var CodeBuddyStopHook []byte
+
 //go:embed openclaw/SKILL.md
 var OpenClawSkill []byte
 
@@ -121,5 +133,5 @@ var HermesCompactHook []byte
 
 // All returns the embedded filesystem for inspection.
 //
-//go:embed claude codex cursor trae qoder qoderwork openclaw nanoclaw nanobot pi hermes
+//go:embed claude codex cursor trae qoder qoderwork codebuddy openclaw nanoclaw nanobot pi hermes
 var All embed.FS

internal/setup/assets/codebuddy/SKILL.md

@@ -0,0 +1,46 @@
+---
+name: mnemon
+description: Persistent memory CLI for CodeBuddy. Store facts, recall past knowledge, link related memories, manage lifecycle.
+---
+
+# mnemon
+
+## Workflow
+
+1. **Remember**: `mnemon remember "<fact>" --cat <cat> --imp <1-5> --entities "e1,e2" --source agent`
+   - Diff is built in: duplicates are skipped, conflicts are auto-replaced.
+   - Output includes `action` (added/updated/skipped), `semantic_candidates`, and `causal_candidates`.
+2. **Link** (evaluate candidates from step 1 using judgment):
+   - Review `causal_candidates`: link only when the memories are genuinely causally related.
+   - Review `semantic_candidates`: high `similarity` alone is not enough; skip unrelated keyword matches.
+   - Syntax: `mnemon link <id> <candidate> --type <causal|semantic> --weight <0-1> [--meta '<json>']`
+3. **Recall**: `mnemon recall "<query>" --limit 10`
+
+## Commands
+
+```bash
+mnemon remember "<fact>" --cat <cat> --imp <1-5> --entities "e1,e2" --source agent
+mnemon link <id1> <id2> --type <type> --weight <0-1> [--meta '<json>']
+mnemon recall "<query>" --limit 10
+mnemon search "<query>" --limit 10
+mnemon import --dry-run <file>
+mnemon import <file>
+mnemon forget <id>
+mnemon related <id> --edge causal
+mnemon gc --threshold 0.4
+mnemon gc --keep <id>
+mnemon status
+mnemon log
+mnemon store list
+mnemon store create <name>
+mnemon store set <name>
+mnemon store remove <name>
+```
+
+## Guardrails
+
+- Use memory only when it can materially improve continuity or task quality.
+- Do not store secrets, passwords, tokens, private keys, or short-lived operational noise.
+- Categories: `preference`, `decision`, `insight`, `fact`, `context`
+- Edge types: `temporal`, `semantic`, `causal`, `entity`
+- Max 8,000 chars per insight.

internal/setup/assets/codebuddy/prime.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+PROMPT_DIR="${MNEMON_DATA_DIR:-$HOME/.mnemon}/prompt"
+if [ ! -f "${PROMPT_DIR}/guide.md" ] && [ -f "${HOME}/.mnemon/prompt/guide.md" ]; then
+  PROMPT_DIR="${HOME}/.mnemon/prompt"
+fi
+
+if ! command -v mnemon >/dev/null 2>&1; then
+  echo "[mnemon] Warning: mnemon not found in PATH."
+  [ -f "${PROMPT_DIR}/guide.md" ] && cat "${PROMPT_DIR}/guide.md"
+  exit 0
+fi
+
+STATS=$(mnemon status 2>/dev/null)
+if [ -n "$STATS" ]; then
+  INSIGHTS=$(echo "$STATS" | sed -n 's/.*"total_insights": *\([0-9]*\).*/\1/p' | head -1)
+  EDGES=$(echo "$STATS" | sed -n 's/.*"edge_count": *\([0-9]*\).*/\1/p' | head -1)
+  echo "[mnemon] Memory active (${INSIGHTS:-0} insights, ${EDGES:-0} edges)."
+else
+  echo "[mnemon] Memory active."
+fi
+
+[ -f "${PROMPT_DIR}/guide.md" ] && cat "${PROMPT_DIR}/guide.md"