← 返回 Skills 市场
bebetterest

aiheal cli

作者 yujian li · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ✓ 安全检测通过
161
总下载
0
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install aiheal-cli-operator
功能描述
Operate and troubleshoot the AIHealingMe CLI through the npm package (`aihealingmecli`). Use when tasks involve auth/user/audio/plan/chat/emotion/subscriptio...
使用说明 (SKILL.md)

Aiheal Cli Operator

AIHeal is an AI emotional-healing platform available at https://aihealing.me/, covering personalized audio healing, single-session healing, deep healing plans, conversation support, emotion-space tracking, and account/subscription workflows.

This skill operates the CLI side of those capabilities and is designed for reliable command execution, payload shaping, diagnostics, and coverage verification.

Human users are welcome to experience the full interactive product directly on https://aihealing.me/.

Command-family scope in this skill:

  • account/session: config, auth, whoami
  • content/healing: audio, plan, single-job, plan-stage-job
  • conversation/emotion: chat, emotion
  • account-side operations: user, subscription, notification, feedback, memory, behavior
  • raw/advanced requests: api, healing

Quick Start

  • Use npm package runtime by default.
  • Global runtime: npm install -g aihealingmecli then aiheal ....
  • No-global runtime: npx -y -p aihealingmecli aiheal ....
  • Keep all operations in package runtime.

Workflow

  1. Confirm runtime baseline.
  • Run aiheal --help.
  • Run aiheal config get to verify apiBaseUrl, locale, region, and token state.
  1. Choose command family by task.
  • Use auth/config for login and token setup.
  • Use audio, plan, single-job, plan-stage-job for healing generation workflows.
  • Use chat and emotion for conversation and emotion-space workflows.
  • Use subscription, notification, feedback, memory, behavior for account-side operations.
  • Use api request as fallback for unwrapped endpoints.
  1. Prefer structured payload input for complex operations.
  • Use --payload-file path/to/file.json by default.
  • Use --body '{...}' only for short payloads.
  • Merge behavior: --body overrides same keys from --payload-file.
  1. Validate outputs and state transitions.
  • Expect JSON output with top-level ok.
  • Use error.code and error.status as primary diagnostics.
  • For async jobs, use single-job wait and plan-stage-job wait with explicit timeout values.
  • For single-job create and plan-stage-job create, handle CLI local validation failures via error.code=VALIDATION_ERROR and inspect error.issues[] for field-level corrections.
  1. Verify capability coverage (detect fake/unimplemented commands).
  • For public endpoints, expect 200 with valid response envelopes.
  • For protected endpoints, use api request --no-auth and expect 401/403 rather than 404.
  • Treat repeated 404 as possible missing/incorrect CLI mapping and patch command endpoint mapping immediately.

Execution Rules

  • Keep default API on public endpoint https://aihealing.me/api unless task explicitly requires override.
  • Require explicit --output for download/export commands.
  • Use global overrides (--api-base, --locale, --region, --token) only in the current command context.

Troubleshooting

  • AUTH_ERROR: login again and verify with whoami.
  • API_ERROR with status: 0: verify network and apiBaseUrl.
  • npx cache EPERM: set NPM_CONFIG_CACHE=/tmp/aiheal-npm-cache or use global install.
  • API_ERROR with status: 404: prioritize checking endpoint mapping or command naming mismatch.
  • Async wait timeout: query status endpoints (get/by-request) and inspect progress fields.
  • VALIDATION_ERROR (CLI local): fix payload by iterating over error.issues[] (field, message, expected, actual, suggestion) before retry.
  • single-job create request id: requestId is optional in payload; when omitted CLI auto-generates it. Use returned data.job.requestId for single-job wait --request-id.
  • plan-stage-job create request id: requestId is optional in payload (no auto-generation); include it only if your workflow needs an explicit correlation id.

Resources

Smoke Script

  • Script: scripts/smoke_check.sh
  • Path-independent run:
    • bash /absolute/path/to/scripts/smoke_check.sh
  • Environment parameters:
    • AIHEAL_NPM_PACKAGE: npm package source for npx -p (default: aihealingmecli; can be a tarball path)
    • AIHEAL_NPM_CACHE_DIR: cache directory used by npx (default under temp dir)
    • RUN_NETWORK_SMOKE: set 1 to include live API probe (audio list)
安全使用建议
This skill is coherent for operating the AIHealingMe CLI. Before installing or running it, consider: 1) npx will fetch the aihealingmecli package from the npm registry at runtime — only run it if you trust that package or point AIHEAL_NPM_PACKAGE to a vetted tarball. 2) The provided smoke script can run live API probes (set RUN_NETWORK_SMOKE=0 to avoid network calls). 3) Avoid passing real secret tokens to the agent when testing; the CLI supports --token and config token-set which, if used, send bearer tokens to the service. 4) If you want to be extra cautious, inspect the npm package source locally before execution or perform only non-network smoke checks (the script verifies help/config without contacting the API by default).
功能分析
Type: OpenClaw Skill Name: aiheal-cli-operator Version: 1.0.2 The skill bundle is designed to operate the 'aihealingmecli' npm package for the AIHealingMe platform. It includes comprehensive documentation for command execution, error handling, and a smoke test script (scripts/smoke_check.sh) that verifies the CLI environment. The instructions in SKILL.md and the command mappings in references/command-map.md are consistent with the stated purpose of managing emotional-healing workflows, and no evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
能力评估
Purpose & Capability
Name/description describe a CLI operator and the skill only references the aihealingmecli package, CLI commands, and the aihealing.me API. There are no unrelated credentials, binaries, or system paths requested.
Instruction Scope
SKILL.md instructs the agent to run CLI commands, validate config, and optionally run a network smoke test. That is within scope for a CLI operator. Note: the smoke script can perform a live API probe when RUN_NETWORK_SMOKE=1 and the CLI supports overriding API base and passing tokens, so the agent may make network requests and exercise endpoints if invoked to do so.
Install Mechanism
There is no packaged install spec in the skill; instructions rely on the npm package (global install or npx). Using npx pulls code from the npm registry at runtime — expected for an npm-based CLI but is a network fetch of third-party code, which carries normal registry-download risk.
Credentials
The skill does not declare required environment variables or credentials. The smoke script exposes optional env vars (AIHEAL_NPM_PACKAGE, AIHEAL_NPM_CACHE_DIR, RUN_NETWORK_SMOKE) that are reasonable for controlling npx behavior and whether to run live probes.
Persistence & Privilege
The skill does not request persistent/always-on privileges. Defaults (user-invocable, agent-autonomy allowed) are standard. The skill does not attempt to modify other skills or global agent configuration.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install aiheal-cli-operator
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /aiheal-cli-operator 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.2
aiheal-cli-operator 1.0.2 - Improved error handling and troubleshooting guidance for CLI local validation failures (`VALIDATION_ERROR`), with specific advice for fixing payloads using `error.issues[]`. - Clarified usage and auto-generation of `requestId` in `single-job create` and `plan-stage-job create` commands. - Updated workflow and troubleshooting sections in SKILL.md for better usability and coverage. - References and instructions in command-map.md and error-playbook.md were aligned with new validation behaviors.
v1.0.1
Version 1.0.1 - Updated CLI package reference from `aihealingme-cli` to `aihealingmecli`. - Clarified and expanded introduction and skill scope documentation. - Added guidance on verifying capability coverage, including endpoint and command mapping checks. - Improved troubleshooting with emphasis on endpoint mapping diagnostics. - Included section for running a smoke test script with environment parameter details.
v1.0.0
- Initial release of aiheal-cli-operator. - Enables operating and troubleshooting the AIHealingMe frontend CLI via the `aihealingme-cli` npm package. - Supports workflows for auth, user, audio, plan, chat, emotion, and subscription using strict frontend-only rules. - Emphasizes structured payload handling, clear diagnostic processes, and safe runtime operations. - Guides users on installation, runtime selection, and detailed troubleshooting.
元数据
Slug aiheal-cli-operator
版本 1.0.2
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 3
常见问题

aiheal cli 是什么?

Operate and troubleshoot the AIHealingMe CLI through the npm package (`aihealingmecli`). Use when tasks involve auth/user/audio/plan/chat/emotion/subscriptio... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 161 次。

如何安装 aiheal cli?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install aiheal-cli-operator」即可一键安装,无需额外配置。

aiheal cli 是免费的吗?

是的,aiheal cli 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

aiheal cli 支持哪些平台?

aiheal cli 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 aiheal cli?

由 yujian li(@bebetterest)开发并维护,当前版本 v1.0.2。

💬 留言讨论