← 返回 Skills 市场
433
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install massive
功能描述
Bash CLI wrapper and OpenClaw skill for Massive's public REST API. Use when Codex or an OpenClaw agent needs to query Massive market-data endpoints from a sh...
安全使用建议
Do not install or enable this skill until the packaging and metadata inconsistencies are resolved. Specifically: (1) ask the publisher to include the scripts/massive executable in the published bundle and to re-run CI/tests so you can inspect the actual script content; (2) confirm the registry metadata lists the credential env vars (MASSIVE_API_KEY_REF / MASSIVE_API_KEY) so installers know what will be required; (3) review the scripts/massive source yourself for any unexpected network endpoints or shell execution; (4) if you will use secret refs that include 'exec', ensure your runtime enforces strict sandboxing and that only trusted principals can supply secret-ref JSON (because exec refs run arbitrary commands); and (5) verify the bundle signature or source repository before trusting it in automated agent workflows.
功能分析
Type: OpenClaw Skill
Name: massive
Version: 1.0.1
The skill includes a high-risk capability allowing arbitrary command execution via `eval "${command}"` when resolving secrets from an `exec` source specified in the `MASSIVE_API_KEY_REF` environment variable (found in `scripts/massive` and documented in `references/openclaw-secrets.md`). This presents a significant Remote Code Execution (RCE) vulnerability if `MASSIVE_API_KEY_REF` can be controlled by untrusted input. While the documentation states this is an intentional feature for OpenClaw alignment and relies on the runtime for sandboxing, the direct use of `eval` without internal sanitization makes it suspicious due to the inherent risk, despite lacking clear evidence of malicious intent within the skill itself.
能力评估
Purpose & Capability
The name/description, SKILL.md, README, and references all describe a Bash CLI (scripts/massive) that resolves OpenClaw secret refs and talks to api.massive.com. However the published bundle in this registry snapshot does not include scripts/massive (the core executable), and the registry metadata at the top lists no required env vars while agents/openai.yaml and the documentation clearly expect MASSIVE_API_KEY_REF / MASSIVE_API_KEY / MASSIVE_BASE_URL. That mismatch suggests a packaging or metadata error and makes it unclear what will actually run if installed as-is.
Instruction Scope
SKILL.md and references restrict network access to the Massive API origin, require redaction, and instruct the agent to resolve secrets via OpenClaw SecretRefs. Those instructions stay within the stated purpose. One noteworthy scope detail: SecretRefs explicitly support an 'exec' source (execute a command and use stdout). That behavior is expected for OpenClaw-style secret resolution but it effectively permits the runtime to execute arbitrary commands when resolving a secret ref, so it increases the trust required of the runtime and any entity that can supply secret-ref payloads.
Install Mechanism
This is instruction-only (no install spec), which is low risk. There are no downloads or extract steps. However the repository bundle as published appears incomplete (scripts/massive is not present in the file manifest despite being required by SKILL.md and tests; BUNDLE_MANIFEST.md explicitly calls this out as a packaging error). That packaging inconsistency should be resolved before trusting the bundle.
Credentials
The environment/credential requirements documented (MASSIVE_API_KEY_REF preferred, MASSIVE_API_KEY fallback, and optional MASSIVE_BASE_URL) are appropriate for a CLI that authenticates to Massive. But the registry metadata at the top claiming 'Required env vars: none' contradicts agents/openai.yaml and the SKILL.md. Additionally, supporting 'exec' secret refs means a provided secret-ref JSON can cause the runtime to run arbitrary commands to fetch secrets — acceptable if the runtime enforces sandboxing, but disproportionate if secret refs can be supplied by untrusted parties or if the runtime has no strong isolation.
Persistence & Privilege
The skill does not request permanent presence (always:false) and does not modify other skills or global agent settings. allow_implicit_invocation is true in agent metadata, which is normal for skills and not a standalone concern here.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install massive - 安装完成后，直接呼叫该 Skill 的名称或使用
/massive触发 - 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
v1.0.1
- Added BUNDLE_MANIFEST.md to define required files for packaged or published skill artifacts.
- Updated documentation to instruct users to include all files listed in BUNDLE_MANIFEST.md when packaging or publishing.
- Expanded agent rules to treat non-api.massive.com URLs as invalid unless MASSIVE_BASE_URL is explicitly set.
v1.0.0
Initial release of the Massive CLI for OpenClaw.
- Provides a Bash CLI wrapper for Massive's public REST API.
- Integrates OpenClaw-style secret reference authentication.
- Includes commands for market-data endpoints: stocks, options, forex, crypto, and indices.
- Enforces security best practices: redacts secrets, logs diagnostics to stderr, enables JSON-first output.
- Supports deterministic REST requests and streamlined CLI agent pipelines.
元数据
常见问题
Massive.com CLI 是什么？
Bash CLI wrapper and OpenClaw skill for Massive's public REST API. Use when Codex or an OpenClaw agent needs to query Massive market-data endpoints from a sh... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 433 次。
如何安装 Massive.com CLI？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install massive」即可一键安装，无需额外配置。
Massive.com CLI 是免费的吗？
是的，Massive.com CLI 完全免费（开源免费），可自由下载、安装和使用。
Massive.com CLI 支持哪些平台？
Massive.com 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 Massive.com CLI？
由 oscraters（@oscraters）开发并维护，当前版本 v1.0.1。
推荐 Skills