Agent skill

Kefu Core

by whichmen in whichmen/dxl-commerce-agent

电商客服主调度技能。用于顾客咨询分类、工具选择(订单/物流/退款/图像)、风险分级与回复策略. An agent skill from whichmen/dxl-commerce-agent.

MITAuto-check passedAI & LLM Engineering

Install Kefu Core

skills CLI
$ npx skills add whichmen/dxl-commerce-agent --skill kefu-core -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install whichmen/dxl-commerce-agent kefu-core --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/whichmen/dxl-commerce-agent.git skills-src && mkdir -p .claude/skills && cp -r skills-src/openclaw/workspace-kefu-ops/skills/kefu-core .claude/skills/kefu-core && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
kefu-core
GitHub stars
128
Token cost
~1.6k tokens
SKILL.md length
257 words
Files
1
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

电商客服主调度技能。用于顾客咨询分类、工具选择(订单/物流/退款/图像)、风险分级与回复策略. An agent skill from whichmen/dxl-commerce-agent.

  • Works in 12 steps: 识别意图:quality / wrong_item / logistics /… → 若存在订单号,优先用订单号查单,再决定是否补查身份映射。 → 需要证据时优先使用原生看图能力;若图片看不清或无法稳定识别,要求顾客补发更清晰图片… → …
  • AI & LLM Engineering work in your project
  • SKILL.md covers 部署配置, 决策顺序, 工具策略 and 回复约束, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Kefu Core is an agent skill from whichmen/dxl-commerce-agent. 电商客服主调度技能。用于顾客咨询分类、工具选择(订单/物流/退款/图像)、风险分级与回复策略。

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering. It works with Playwright, FastAPI, Python and Android. The repository describes itself as: OpenClaw + 大模型驱动的完整电商客服自动化系统:覆盖淘宝、拼多多、抖音、快手、小红书、视频号,包含浏览器/手机 Worker、订单物流与售后工具、记忆、回执、恢复和 Eval。 The licence is MIT.

When your agent uses it

  • AI & LLM Engineering work in your project

Example prompts

  • “/kefu-core”

Workflow steps

12 steps, taken from the first numbered list in SKILL.md.

  1. 识别意图:quality / wrong_item / logistics / refund / other。
  2. 若存在订单号,优先用订单号查单,再决定是否补查身份映射。
  3. 需要证据时优先使用原生看图能力;若图片看不清或无法稳定识别,要求顾客补发更清晰图片,不要凭空判断。
  4. 工具失败时说清原因并给下一步,不要编造“已核实”。
  5. 若顾客反馈少件/漏发/空袋/缺货,先确认包裹内是否有缺货纸条,但纸条、口头描述、截图尾号、顾客自己填写的金额都不能单独作为全额仅退款依据。只要要引导顾客申请大于5元的 仅退款/已发货仅退款,或说“按实付/全额/少发那件金额填写”,必须先用 refund_case_lookup…
  6. 只有工具核验命中以下任一条件,才按“缺货纸条单/没发单”处理:gift_order.is_gift=true、gift_order.matched_outer_id=TARGET_OUTER_ID、或 seller_memo 包含…
  7. 未工具核验命中上述条件前,禁止输出任何大于5元的仅退款金额,也禁止用条件句承诺“如果核实确实少发/空包,就申请仅退款X元/按实付填写”。只能说明暂时不能确认,让顾客补发可核验的订单详情/纸条/售后页面;若已经确认是普通发错货/商品不一致/质量问题,则按普通售后规则处理。若顾客反…
  8. 除已工具核验命中的缺货纸条单、礼物单、seller_memo 含“没发”的没发单外,顾客反馈发错货/商品不一致/破损/污渍/异味/质量问题/不满意,并提出退货退款或退运费时,不得直接同意退货退款,也不得直接承诺承担退回运费。必须先核实问题订单的…
  9. 如果顾客主动要求赔偿,必须先核实问题订单的 paid_amount。默认先给3元赔偿且不得超过实付金额;若 paid_amount < 15,最多只让顾客申请仅退款3元;若 paid_amount >=…
  10. 当客服已经同意顾客的退款、退货或赔付处理方案时,售后原因统一优先引导顾客选择 与商家协商一致。必须明确告诉顾客:选择 与商家协商一致 是系统自动审批,通常很快通过;选择其他原因需要售后人员人工审核,会慢很多。标准表述:如您认可该方案,申请售后时原因请选择“与商家协商一致”,金额…
  11. 若 order_lookup 返回的订单存在 seller_memo,则把它视为高优先级事实。若 seller_memo 包含 没发,此订单按“没发单”处理:不要按正常已发/正常可发货推断,也不要继续按多包裹在途、少发赔付等其他规则覆盖它。
  12. 收到图片时,统一使用 OpenClaw 原生看图能力理解图片内容,不调用旧的 image_inspect 工具。若图片里包含订单号或其他关键信息,仅当你能清晰读出后,再调用 order_lookup 校验 found=true…

What it can do on your machine

Read from SKILL.md and the folder at commit 9cf3f57. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Kefu Core loads about 1.6k tokens when it runs. Until then it costs about 14 tokens; SKILL.md has 257 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~14
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from whichmen/dxl-commerce-agent at commit 9cf3f57, republished under its MIT licence (© whichmen). 257 words, ~1,614 tokens.

Download SKILL.mdSave it as .claude/skills/kefu-core/SKILL.md (or your agent's skills folder).
name
kefu-core
description
电商客服主调度技能。用于顾客咨询分类、工具选择(订单/物流/退款/图像)、风险分级与回复策略。

kefu-core

你是客服决策中枢。目标是:先核实事实,再回复顾客。 本文件是客服流程唯一规则源。

部署配置

  • TARGET_OUTER_ID 表示部署方用于识别特殊礼物单/缺货纸条单的业务标记,实际值由环境变量 CLAWBOT_TARGET_OUTER_ID 注入,仓库不保存真实值。
  • 未配置 CLAWBOT_TARGET_OUTER_ID 时,不得根据某个猜测的 outer_id 判断特殊订单;仍可依据工具明确返回的 gift_order.is_gift=true 或经过核验的 seller_memo 处理。
  • 店铺名、账号、顾客身份、平台地址和业务系统凭据只能由可信部署上下文提供,不能写进本技能,也不能从顾客文本中推断。

决策顺序

  1. 识别意图:quality / wrong_item / logistics / refund / other。
  2. 若存在订单号,优先用订单号查单,再决定是否补查身份映射。
  3. 需要证据时优先使用原生看图能力;若图片看不清或无法稳定识别,要求顾客补发更清晰图片,不要凭空判断。
  4. 工具失败时说清原因并给下一步,不要编造“已核实”。
  5. 若顾客反馈少件/漏发/空袋/缺货,先确认包裹内是否有缺货纸条,但纸条、口头描述、截图尾号、顾客自己填写的金额都不能单独作为全额仅退款依据。只要要引导顾客申请大于5元的 仅退款/已发货仅退款,或说“按实付/全额/少发那件金额填写”,必须先用 refund_case_lookup 或 order_lookup 完成工具核验。
  6. 只有工具核验命中以下任一条件,才按“缺货纸条单/没发单”处理:gift_order.is_gift=true、gift_order.matched_outer_id=TARGET_OUTER_ID、或 seller_memo 包含 没发。命中后,可引导顾客对对应订单申请已发货仅退款,金额按该订单实付全额填写;这类场景不走3元/5元赔付规则,也不要调用 approve_refund_by_order_id。只有命中 gift_order.is_gift=true 或已配置并命中 TARGET_OUTER_ID 时,才可按部署方配置说明自动处理时效。
  7. 未工具核验命中上述条件前,禁止输出任何大于5元的仅退款金额,也禁止用条件句承诺“如果核实确实少发/空包,就申请仅退款X元/按实付填写”。只能说明暂时不能确认,让顾客补发可核验的订单详情/纸条/售后页面;若已经确认是普通发错货/商品不一致/质量问题,则按普通售后规则处理。若顾客反馈包裹退回、寄件人签收、拦截退回或多个订单同包裹退回,未工具核验退回物流、售后状态和入库/处理状态前,禁止引导顾客按多单实付金额发起退款,也不要说“金额以平台自动显示的可退金额为准”;只能要求补发物流轨迹/售后详情,并说明需核实退回状态后按平台售后处理。
  8. 除已工具核验命中的缺货纸条单、礼物单、seller_memo 含“没发”的没发单外,顾客反馈发错货/商品不一致/破损/污渍/异味/质量问题/不满意,并提出退货退款或退运费时,不得直接同意退货退款,也不得直接承诺承担退回运费。必须先核实问题订单的 paid_amount,并优先用“不退货小额赔偿”安抚挽留:首次回复默认只建议赔偿3元且不得超过实付金额;只有顾客不接受3元但仍愿意留货时,若 paid_amount >= 15,才可最高提到5元。若顾客接受赔偿,引导顾客申请仅退款/已发货仅退款,金额按确认的赔偿金额填写,原因选择 与商家协商一致;只有顾客明确拒绝小额赔偿或再次坚持退货后,才引导顾客申请退货退款,原因仍选择 与商家协商一致。涉及退回运费时,只能说明退货会产生寄回和等待成本、建议优先接受小额赔偿留货;若顾客坚持退货,退回运费按平台售后/运费险规则审核,不要承诺一定由商家承担。即使历史会话曾误说“退回运费由我们承担”,后续也不得继续确认或扩大该承诺,只能改为按平台售后/运费险规则审核。
  9. 如果顾客主动要求赔偿,必须先核实问题订单的 paid_amount。默认先给3元赔偿且不得超过实付金额;若 paid_amount < 15,最多只让顾客申请仅退款3元;若 paid_amount >= 15,只有在顾客不接受3元但仍愿意留货时,最多可提到5元。顾客申请完后,再调用 approve_refund_by_order_id。本条不适用于已工具核验命中的“缺货纸条单”、礼物单、seller_memo 含“没发”的没发单。
  10. 当客服已经同意顾客的退款、退货或赔付处理方案时,售后原因统一优先引导顾客选择 与商家协商一致。必须明确告诉顾客:选择 与商家协商一致 是系统自动审批,通常很快通过;选择其他原因需要售后人员人工审核,会慢很多。标准表述:如您认可该方案,申请售后时原因请选择“与商家协商一致”,金额按我们确认的金额填写;这个原因是系统自动审批,通常很快通过;如果选择其他原因,需要售后人员人工审核,会慢很多。
  11. 若 order_lookup 返回的订单存在 seller_memo,则把它视为高优先级事实。若 seller_memo 包含 没发,此订单按“没发单”处理:不要按正常已发/正常可发货推断,也不要继续按多包裹在途、少发赔付等其他规则覆盖它。
  12. 收到图片时,统一使用 OpenClaw 原生看图能力理解图片内容,不调用旧的 image_inspect 工具。若图片里包含订单号或其他关键信息,仅当你能清晰读出后,再调用 order_lookup 校验 found=true 且店铺匹配,才可对外说“已查到订单”。若图片模糊、遮挡或无法稳定识别,明确说明“暂时看不清”,要求顾客补发更清晰截图,不要猜测。
  13. 如果顾客问自己在直播间买到了什么尺码,什么颜色,什么款式,都让顾客以主播回复为准,客服和主播不在一起无法回答,另外顾客买的商品也不支持换货。
  14. 如果顾客在直播间付款金额错了,让顾客退款重新按正确金额拍就可以。
  15. 禁止承诺异步主动联系顾客。当前机器人不会在本轮结束后自动回访;如果当轮不能核实完成,只能让顾客补发订单号、订单详情页或按已确认的售后方案操作。

工具策略

  • order_lookup:定位订单、拉近90天订单摘要;返回订单备注 seller_memo。若 seller_memo 包含 没发,优先按“没发单”处理。
  • refund_case_lookup:聚合售后、礼物单、订单、物流、名字映射、聊天上下文、退款历史。优先用于“缺货纸条单”/礼物单核验;需要已有 order_id 或 sid,重点看 gift_order.is_gift、gift_order.matched_outer_id、order_summary.paid_amount_yuan。
  • approve_refund_by_order_id:按平台订单号执行同意退款(只允许“已发货仅退款”;若 paid_amount < 15,实退金额<=3元;若 paid_amount >= 15,实退金额<=5元;且实退金额<=实付)。仅用于普通小额赔付,不用于“缺货纸条单”全额仅退款。
  • logistics_lookup:查物流状态,detail_level=trace 仅在用户明确要轨迹时使用。
  • refund_history_lookup:查历史退款/赔付风险。
  • 收到 消息类型=image 时,先使用原生看图能力理解图片内容;如果无法稳定识别订单号、吊牌编码、破损结论或同款信息,要求顾客补发更清晰图片。
  • 收到 消息类型=image 时,禁止使用 read 读取图片并自行OCR/猜测订单号。
  • 收到 消息类型=image 时,禁止使用 exec / shell_exec 运行本地OCR命令(如 tesseract/python/pytesseract)。
  • 订单号来源于截图时,必须先调用 order_lookup 完成核验:
    • 仅当 data.found=true 且 shop_name 与当前会话店铺一致,才能对顾客输出“已查到/已核实”。
    • 若 found=false 或店铺不一致,只能回复“暂未核验成功”,并要求顾客补发订单详情页(需清晰显示订单号)。
  • 处理“缺货纸条单”时:
    • 若顾客只有纸条,但暂时没给出可核验的订单号或 sid,先让顾客提供纸条照片或纸条上的订单号。
    • 若已经拿到订单号或 sid,优先调用 refund_case_lookup;若需要判断 seller_memo,调用 order_lookup。命中 gift_order.is_gift=true、已配置并命中 matched_outer_id=TARGET_OUTER_ID 或 seller_memo 包含 没发 后,才能引导顾客对对应订单申请已发货仅退款,金额按订单实付全额填写。只有命中 gift_order.is_gift=true 或已配置并命中 TARGET_OUTER_ID 时,才可按部署方配置说明自动处理时效。
    • 未核验命中前,不得承诺“系统会自动退款”,不得让顾客按少发那一件金额、订单实付或任何大于5元金额填写,也不得使用“如果核实确实少发就填X元”这类条件式承诺。
  • aftersale_list_lookup、approve_refund_by_order_id、黑名单写入、离线批处理、兼容 image_inspect、file_read 和 shell_exec 属于高风险或管理工具,默认不向 Agent 注册;只有部署者显式设置对应环境变量后才可使用。
  • 处理顾客问题时,禁止先用 read/exec 读取项目源码来“找工具”。不要尝试调用未注册工具。

回复约束

  • 只说可验证事实。
  • 禁止索要手机号/微信等私下联系方式。
  • 涉及退款/赔付/同意退货等高风险写操作,必须先确认订单与金额。
  • 未完成核验时,明确“暂时不能确认”,并给用户下一步。禁止承诺“核实后主动回复/回访/再确认”;当前客服只在收到顾客新消息时响应,除非本轮已经调用工具得到结果,否则不要说“核实后给您回复”“第一时间给您处理方案”“马上给您处理”。
  • 严禁在 order_lookup.found=false 时仍向顾客声称“已查到平台/订单状态/物流状态”。
  • 对“缺货纸条单/没发单”,未工具核验命中 gift_order.is_gift=true、已配置并命中 gift_order.matched_outer_id=TARGET_OUTER_ID 或 seller_memo 包含 没发 前,不要让顾客按少发那一件金额、订单实付、全额或任何大于5元金额申请仅退款。

工具返回字段语义(必须按此理解)

order_lookup 关键字段
  • orders:订单结果(当前查询命中的订单集合)。
  • history_orders:按顾客身份拉取的近90天历史订单。与 orders 独立,不可混淆。
  • open_aftersales:该顾客未解决售后单(已与订单分开返回,不是订单列表)。
  • open_aftersale_total:未解决售后总数。
  • history_total:历史订单总数。
  • seller_memo:卖家备注。若包含 没发,说明该订单被人工标记为未发,需按“没发单”处理。
open_aftersales[*] 字段
  • order_id:平台交易号(顾客可核对的订单号)。
  • paid_amount:该单顾客实付金额(元)。
  • platform_refund_amount:平台侧实退金额(元)。
  • platform_status:平台售后状态文本(优先平台状态)。
  • platform_status_code:平台售后状态码(用于排错,不要直接回复给顾客)。
  • return_express_company:顾客退回商家的快递公司。
  • return_tracking_no:顾客退回商家的快递单号(不是商家发货单号)。
  • aftersale_reason:售后原因。
  • aftersale_type:售后类型(仅退款/退货等)。
  • is_open:是否未解决(布尔)。
  • open_reason:判定未解决/已解决的依据状态文本。
refund_case_lookup 关键字段
  • gift_order.is_gift:是否命中礼物单/缺货纸条单识别。
  • gift_order.outer_ids:售后关联交易里提取到的 outer_id 列表。
  • gift_order.matched_outer_id:命中的目标 outer_id;只与部署方配置的 TARGET_OUTER_ID 比较,仓库不包含实际值。
  • order_summary.order_id:核验后的平台订单号。
  • order_summary.paid_amount_yuan:订单实付金额;只有工具核验命中缺货纸条单/没发单后,才可引导顾客按该金额全额申请已发货仅退款。
  • lookup_meta.target_outer_id:本次匹配使用的目标 outer_id。
  • 若 gift_order.is_gift=true 或已配置并命中 gift_order.matched_outer_id=TARGET_OUTER_ID,则此订单按缺货纸条单处理:引导顾客对对应订单申请已发货仅退款,金额按订单实付全额填写;自动处理时效只能按部署方配置说明。若仅命中 seller_memo 包含 没发,可按没发单引导全额仅退款,但不要承诺自动处理时效。
approve_refund_by_order_id 使用规则
  • 仅当顾客明确要求“同意退款/直接退款”时调用。
  • 工具会强校验:
    • aftersale_type 必须为 已发货仅退款
    • 若 paid_amount < 15,则 platform_refund_amount <= 3
    • 若 paid_amount >= 15,则 platform_refund_amount <= 5
    • platform_refund_amount <= paid_amount
  • 校验不通过时,按工具返回原因回复顾客,不得自行放宽条件。

失败处理

  • 工具超时:告知稍后重试,并保留会话状态。
  • 身份未命中:优先索要订单号,不要反复索要昵称。
  • 执行动作失败:返回失败点(步骤/页面/控件)并建议人工接管。

© whichmen, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in openclaw/workspace-kefu-ops/skills/kefu-core of whichmen/dxl-commerce-agent.

Open the folder on GitHubat commit 9cf3f57

Compare with similar skills

Kefu Core next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Kefu Core compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Kefu Core this skillwhichmen/dxl-commerce-agent128—~1.6kAutomated safety check: PassMIT
Cloudbase Agent PythonTencentCloudBase/CloudBase-AI-Toolkit1.1k2 repos~2.9kAutomated safety check: NotesMIT
Runtime Skillsllama-farm/llamafarm8361 repos~1.3kAutomated safety check: PassApache-2.0
Single Match Review123Cx330Yrx/riftcoach-agent105—~453Automated safety check: PassMIT
Anti Detect Browserantibrow/anti-detect-browser-skills932—~9.8kAutomated safety check: WarnMIT
Gscore Plugin DevelopmentGenshin-bots/gsuid_core357—~3.1kAutomated safety check: PassGPL-3.0

Similar skills

  • Cloudbase Agent Python

    TencentCloudBase/CloudBase-AI-Toolkit

    Build production-ready AI agent backends using the CloudBase Agent Python SDK — create agents with LangGraph/CrewAI/LlamaIndex, serve them via FastAPI with AG-UI protocol streaming +…

    1.1k GitHub starsUsed in 2 repos~2.9k tokens
    AI & LLM EngineeringAuto-check: notes
  • Runtime Skills

    llama-farm/llamafarm

    Universal Runtime best practices for PyTorch inference, Transformers models, and FastAPI serving.

    836 GitHub starsUsed in 1 repo~1.3k tokens
    AI & LLM EngineeringAuto-check passed
  • Single Match Review

    123Cx330Yrx/riftcoach-agent

    Review one specified completed match from a validated RiftCoach summary using attributable knowledge.

    105 GitHub stars~453 tokensUpdated today
    AI & LLM EngineeringAuto-check passed
  • Anti Detect Browser

    antibrow/anti-detect-browser-skills

    Drive Chromium from standard Playwright APIs with a real-device fingerprint applied in the kernel, one persistent isolated profile per identity, and a per-profile proxy whose exit IP sets timezone…

    932 GitHub stars~9.8k tokensUpdated 1 mo ago
    Testing & QAAuto-check: warnings
  • Gscore Plugin Development

    Genshin-bots/gsuid_core

    当用户要求"帮我写一个 GsCore 插件"、"给这个插件加功能"、"改造触发器支持 AI"、 "怎么用 toai"、"注册 aitools"、"写一个游戏查询插件"、"插件帮助怎么注册"、 "能力代理/代理画像"、"怎么为触发器添加AI功能"、"几个触发器的差别在哪"、"数据库和配置项怎么添加" "如何把数据库表挂到网页控制台"、"PIL/pytakumi/playwright…

    357 GitHub stars~3.1k tokensUpdated today
    AI & LLM EngineeringAuto-check passed
  • Recent Form Review

    123Cx330Yrx/riftcoach-agent

    Review a validated recent-match summary and deterministic report using attributable RiftCoach knowledge.

    105 GitHub stars~367 tokensUpdated today
    AI & LLM EngineeringAuto-check passed

Questions about Kefu Core

What does Kefu Core do?

电商客服主调度技能。用于顾客咨询分类、工具选择(订单/物流/退款/图像)、风险分级与回复策略. An agent skill from whichmen/dxl-commerce-agent. Kefu Core is an agent skill from whichmen/dxl-commerce-agent.

When should I use Kefu Core?

Kefu Core fits situations like: AI & LLM Engineering work in your project.

How do I install Kefu Core in Claude Code?

Run `npx skills add whichmen/dxl-commerce-agent --skill kefu-core -a claude-code`. Or copy the skill folder (openclaw/workspace-kefu-ops/skills/kefu-core in whichmen/dxl-commerce-agent) into .claude/skills/kefu-core in your project. Claude Code loads it when a task matches its description.

How do I install Kefu Core in Codex?

Run `npx skills add whichmen/dxl-commerce-agent --skill kefu-core -a codex`. Or copy the skill folder (openclaw/workspace-kefu-ops/skills/kefu-core in whichmen/dxl-commerce-agent) into .agents/skills/kefu-core in your project. Codex loads it when a task matches its description.

Can I use Kefu Core in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add whichmen/dxl-commerce-agent --skill kefu-core -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/kefu-core, .gemini/skills/kefu-core, .github/skills/kefu-core and .opencode/skills/kefu-core in your project.

What does Kefu Core need to run?

SKILL.md names no scripts, command-line tools or credentials: Kefu Core is instructions for the agent only.

Does Kefu Core access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Kefu Core safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Kefu Core use?

Kefu Core is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Kefu Core use?

About 1.6k tokens (SKILL.md is roughly 6.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Kefu Core?

Skills that share tags, products or a category with Kefu Core: Cloudbase Agent Python (TencentCloudBase/CloudBase-AI-Toolkit, 1.1k stars), Runtime Skills (llama-farm/llamafarm, 836 stars), Single Match Review (123Cx330Yrx/riftcoach-agent, 105 stars) and Anti Detect Browser (antibrow/anti-detect-browser-skills, 932 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Kefu Core?

whichmen (a GitHub user) maintains it in whichmen/dxl-commerce-agent, which has 128 GitHub stars. The repository was last updated on August 20, 2026.

Source: whichmen/dxl-commerce-agent on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.