D
Warning
45/100
feishu-claude-code
cc-lark — a multi-agent fleet platform that uses your IM group as the bus. Fan out 7 parallel sub-agents across 6 model backends, cross-model cosign, runtime MCP that outlives the turn. 飞书 / Lark / Telegram × 多 Agent 舰队。
Warning. Serious findings were identified. Review the full report before connecting. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
⚡ Live Monitored: not connected
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
yixin-1024
Source: github_repo_search
Is this your MCP?
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
// key findings
🚨
Reads files from sensitive locations
Touches: ~/.claude/.credentials.json, ~/.claude/.credentials.json
🔐
You'll be asked for 9 credentials: CC_LARK_CONTROL_TOKEN, OPENCODE_API_KEY, MIMO_API_KEY, GROK_API_KEY, MAKA_API_KEY, AGY_API_KEY, GEMINI_API_KEY, FEISHU_APP_SECRET, FEISHU_VERIFICATION_TOKEN
These are read from process.env at runtime. Make sure you trust where they’ll be sent.
// environment variables
To run this server yourself, you supply these values. They go in your own MCP client configuration and stay on your machine. The secret label means the value is sensitive, not that the server mishandles it.
🔐 secret
CC_LARK_CONTROL_TOKEN控制面 Bearer token；默认保存在 ~/.feishu-claude/control-token（0600）config
CC_LARK_CONTROL_PORTconfig
CC_TG_SET_COMMANDSbot 启动时会自动 setMyCommands 注册斜杠命令菜单；想自己在 BotFather 里维护就设 CC_TG_SET_COMMANDS=0config
CC_TG_ASR_PROFILE借同进程 Lark profile 的 ASR（CC_TG_ASR_PROFILE），没有 Lark profile 时明确报错config
QUOTA_NOTIFY_PROFILEconfig
QUOTA_NOTIFY_OPEN_IDconfig
QUOTA_WATCH_INTERVAL_SECconfig
ACCOUNT_AUTO_SWITCH撞 Claude Max 用量墙时一次性紧急切到有余量的 saved 账户并 resume 续跑（不依赖 ACCOUNT_AUTO_SWITCH）config
ACCOUNT_SWITCH_COOLDOWN_SECconfig
CODEX_QUOTA_WATCH_INTERVAL_SECconfig
CODEX_QUOTA_NOTIFY_PROFILEconfig
CODEX_QUOTA_NOTIFY_OPEN_IDconfig
CC_LARK_HTTP_PORTconfig
CC_LARK_CALLBACK_PORTconfig
CC_LARK_CONTROL_TOKEN_FILEconfig
CC_LARK_CHAT_IDconfig
CC_LARK_THREAD_IDconfig
CC_LARK_ANCHORconfig
CC_LARK_USER_IDprompt cache 友好：注入的 Lark 系统提示在同一话题内逐轮字节一致（本轮消息 id / 提问者走用户消息开头的【本轮】行 + CC_LARK_MESSAGE_ID / env），续轮首调命中整段历史缓存，而不是每轮重写config
CC_LARK_PROFILEconfig
CC_LARK_MESSAGE_IDprompt cache 友好：注入的 Lark 系统提示在同一话题内逐轮字节一致（本轮消息 id / 提问者走用户消息开头的【本轮】行 + / CC_LARK_USER_ID env），续轮首调命中整段历史缓存，而不是每轮重写config
CC_LARK_HTTP_MAX_BODYconfig
CC_LARK_HTTP_READ_TIMEOUT_SECconfig
NGROK_DOMAINngrok 固定域名 (避免每次重启换 URL)config
CC_LARK_CRED_BACKEND取不到）可以用 =file 强制走文件模式。config
ACCOUNT_SWITCH_ON_LIMIT撞 Claude Max 用量墙自动兜底：一轮在 API 上撞 "You've hit your session limit" → 自动切到有余量的 saved 账户并 resume 同一 session 续跑（，与主动切换 ACCOUNT_AUTO_SWITCH 独立）；没有可切的账户 → 在话题里排一个到配额重置时刻的自动唤醒，醒来自动『继续』（CC_LARK_WAKE_ON_LIMIT）config
AGY_BOT_HOMEconfig
AGY_DEFAULT_MODELconfig
AGY_EFFORTconfig
CC_LARK_WAKE_MCPMCP server。设置 =0 可关闭自动注入。config
PYTEST_CURRENT_TESTconfig
CODEX_DANGEROUS_BYPASSconfig
CODEX_IDLE_TIMEOUT_SECconfig
OPENCODE_DANGEROUS_SKIPconfig
OPENCODE_IDLE_TIMEOUT_SECconfig
MIMO_DANGEROUS_SKIPconfig
MIMO_IDLE_TIMEOUT_SECconfig
GROK_DANGEROUS_SKIPconfig
GROK_IDLE_TIMEOUT_SECconfig
GROK_MAX_TURNSconfig
MAKA_DANGEROUS_SKIPconfig
MAKA_IDLE_TIMEOUT_SECconfig
MAKA_MAX_STEPSconfig
AGY_DANGEROUS_SKIPconfig
AGY_IDLE_TIMEOUT_SECconfig
DEFAULT_MODEL默认 Claude 模型；可用 CLI 别名 opus[1m]/sonnet[1m]/fable[1m]/haiku（=系列最新）或钉死具体版本config
DISPATCH_MODELconfig
CODEX_BINconfig
CODEX_SANDBOX_MODEconfig
CODEX_APPROVAL_POLICYconfig
OPENCODE_BINconfig
OPENCODE_PROVIDER🔐 secret
OPENCODE_API_KEYconfig
OPENCODE_API_KEY_ENVconfig
MIMO_BINconfig
MIMO_PROVIDERconfig
MIMO_BASE_URL🔐 secret
MIMO_API_KEYconfig
MIMO_VARIANTconfig
GROK_BINconfig
GROK_HOME_DIR🔐 secret
GROK_API_KEYconfig
GROK_API_KEY_ENVconfig
MAKA_BINconfig
MAKA_WORKSPACE_ROOTconfig
MAKA_CONNECTION🔐 secret
MAKA_API_KEYconfig
MAKA_API_KEY_ENVconfig
MAKA_BASE_URLconfig
MAKA_BASE_URL_ENVconfig
AGY_BINconfig
AGY_MODEL_PROVIDER🔐 secret
AGY_API_KEY🔐 secret
GEMINI_API_KEYconfig
AGY_API_KEY_ENVconfig
AGY_PROXYconfig
AGY_PRINT_TIMEOUTconfig
YUSHITAI_OPEN_IDconfig
ZHONGSHU_OPEN_IDconfig
MENXIA_OPEN_IDconfig
SHANGSHU_OPEN_IDconfig
GANHUODE_OPEN_IDconfig
BOSS_OPEN_IDconfig
FEISHU_APP_ID飞书/Lark 应用 App ID🔐 secret
FEISHU_APP_SECRET飞书/Lark 应用 App Secretconfig
LARK_PLATFORMconfig
LEGACY_PROFILE_NAMEconfig
DEFAULT_CWDClaude CLI 默认工作目录config
RUNNERconfig
ALLOWED_OPEN_IDS访问控制： 用户白名单、ALLOWED_GROUP_CHAT_IDS 群聊白名单，未授权者静默忽略config
ALLOWED_GROUP_CHAT_IDS访问控制：ALLOWED_OPEN_IDS 用户白名单、 群聊白名单，未授权者静默忽略🔐 secret
FEISHU_VERIFICATION_TOKENconfig
LARK_CLI_PROFILEconfig
CLAUDE_ENV_FILEconfig
CLAUDE_RUNNERconfig
PROFILESconfig
ENABLE_TRINITY何时启用： 设 =true 并给 profile 配 <NAME>_ROLE=yushitai/zhongshu/menxia/shangshu/ganhuode。两个条件都满足才生效——这样允许 .env 里同时保留 trinity 配置和遗留 profile，开关切换。config
THREAD_SHARED_SESSIONconfig
CLAUDE_CLI_PATHClaude CLI 可执行文件路径config
PERMISSION_MODE工具权限模式config
CALLBACK_PORT1. 使用 ngrok 暴露本机 （默认 9981）config
CONTROL_PORTngrok 只能指向 CALLBACK_PORT。不要暴露 ；后者仅供本机 MCP/运维调用。config
STREAM_CHUNK_SIZE流式推送的字符积累阈值config
CC_LARK_API_TIMEOUTconfig
CC_LARK_CARD_ACTION_TTL_SECconfig
CLAUDE_PTY_FORCE_SESSION_IDconfig
CLAUDE_EFFORTconfig
CC_LARK_DISALLOWED_TOOLSconfig
CODEX_AUTO_CONTINUEconfig
CODEX_REASONING_EFFORTconfig
CODEX_MAX_CONTINUEconfig
CODEX_CONTINUE_BUDGET_SECconfig
GROK_EFFORTconfig
CC_LARK_LAUNCHD_LABELconfig
CC_LARK_TASKS_YAMLconfig
CC_LARK_WRITE_OP_MARKERSconfig
CC_LARK_SAFEGUARDS_FALLBACK_MODELconfig
CODEX_MODEL_FALLBACKcodex 模型被当前账户拒绝（400 "model is not supported…"）时本轮改用的模型；空串关闭config
DISPATCH_CONCURRENCY_CAP（默认 7，触顶也回 429，稍后重试即可）。派进来的子会话同样config
DISPATCH_WAVE_DEBOUNCE_SECconfig
CC_LARK_API_MAX_PROMPT_CHARSconfig
CC_LARK_API_MAX_VAR_CHARSconfig
CC_LARK_API_SIG_SKEW_SECconfig
CC_LARK_API_IDEMPOTENCY_TTL_SECconfig
CC_LARK_API_TASK_TTL_SECconfig
CC_LARK_API_DISPATCH_TIMEOUT_SECconfig
CC_LARK_API_READ_TIMEOUT_SECconfig
CC_LARK_API_STATE_FILEconfig
CC_LARK_API_TRUSTED_PROXY_IPSconfig
LARK_NO_RETRY_CODESconfig
LARK_CARD_MODEconfig
LARK_CARD_STREAMINGconfig
LARK_CARD_SCHEMAconfig
LARK_CARD_FINAL_CONFIRM_DELAYconfig
CC_LARK_GROK_DISALLOWED_TOOLSconfig
CLAUDE_MODELconfig
CC_LARK_HANDOVER_DIR时可重读， 可改目录）；取消原话题所有待触发的 wake_me_in（否则config
XDG_CONFIG_HOMEconfig
CC_LARK_RESUME_STOREconfig
CC_LARK_LOG_PATHbot 日志 copytruncate 轮转：超过阈值 MB 就轮转，保留 N 代；CC_LARK_LOG_MAX_MB=0 关闭config
CC_LARK_LOG_MAX_MB日志轮转：bot 日志超过 （默认 50）自动 copytruncate 轮转，保留 CC_LARK_LOG_KEEP 代config
CC_LARK_LOG_KEEP日志轮转：bot 日志超过 CC_LARK_LOG_MAX_MB（默认 50）自动 copytruncate 轮转，保留 代config
CC_LARK_LOCK_GLOBSconfig
SCHED_MENTION_OWNERconfig
CC_LARK_WAKE_STOREconfig
CC_LARK_TASK_ROUTESconfig
CC_TG_API_ROOTconfig
CC_TG_EDIT_MIN_INTERVAL流式过程中的中间帧会按 合并（撞不到 Telegram 的 floodconfig
CC_TG_HTTP_TIMEOUTconfig
CC_TG_EDIT_TIMEOUTconfig
CC_TG_FINAL_CONFIRM_DELAYconfig
CC_TG_CALLBACK_TTL_SECconfig
CC_TG_CALLBACK_DEDUP_SECconfig
CC_TG_ALBUM_WINDOW_SEC一次发多张图（相册）会被合并成一条消息再交给 agent（攒config
CC_TG_CONFLICT_ESCALATEconfig
CC_TG_POLL_GAP_SECconfig
CC_TG_CONFLICT_MAX_BACKOFFconfig
CC_TG_RECLAIM_AFTERconfig
CC_TG_ALLOWLIST_DIRconfig
CC_TG_BUFFER_DIRconfig
KNOWN_BOT_NAMES// quality suggestions
Tools detected
No tools extracted — insufficient content to verify at Tier 2
No access to sensitive paths
Reads sensitive paths: ~/.claude/.credentials.json, ~/.claude/.credentials.json
Remove reads of sensitive system paths. If you genuinely need them, document why in the README.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
// full audit trail
The findings above are the summary. The full trail, every check we ran, each deduction, the network hosts observed and the dependency advisories, goes to verified publishers, along with an alert whenever a new one lands. Verified publishers can also review each finding and dispute it in one click. Publisher corrections have sharpened several of our checks this month, because the maintainer knows the codebase better than any scanner.
// improvement guidance — verified publishers only
We have 1 concrete improvement we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.
// embed badge in your README
[](https://m8ven.ai/mcp/yixin-1024/feishu-claude-code)Shows verification status without the grade. Want the grade badge instead? Remove
?variant=verified from the URL.commit: 9e3c3eb5ef396b38875cc3d2651de8f6beda3387
code hash: b1693d2bee73e828567c37b76ccce8fa6954a56a3e5a66fedad0ccb97edd9efd
Check MCPs from inside your assistant
Tool Check · MCPVetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-checkcheck_toolsearch_toolscompare_toolsrecommend_alternativescheck_publisherreport_concern
How to add it →Free · no account needed · works in any MCP client