CMIC Skill Scanner (Linux ARM64)

使用内置 Rust 引擎审计待安装的 skill 包或归档，并可选桥接外部 scanner。

Security Scan

The skill's stated purpose (local skill scanner) matches its instructions, but there are packaging/information inconsistencies and optional upload functionality that you should verify before running anything.

Purpose & Capability

Name/description, SKILL.md, and INSTALL.md all describe a local Rust-based scanner; that aligns with the declared purpose. However SKILL.md and INSTALL.md refer to a bundled binary at assets/bin/skillscan, while the registry file manifest does not list that binary — an inconsistency that should be resolved before trusting a claimed prebuilt binary.

Instruction Scope

Runtime instructions are narrowly scoped to scanning local skill packages and saving local reports. They explicitly state network/upload features are disabled by default and that credentials are not accessed unless configured. Still, the docs reference optional flags (--upload-url, --engine external) that, if enabled, will transmit structured findings or run an external engine; users must explicitly opt in to those behaviors.

Install Mechanism

The package is instruction-only (no install spec), which is low-risk. But the documentation advertises a bundled precompiled binary (with a SHA-256), yet that binary file is not present in the manifest. If a binary were included or you download one later, that would be higher risk — prebuilt binaries deserve checksum verification and ideally building from source. The referenced source repo (gitee.com) is plausible but is an external network dependency.

Credentials

The skill declares no required env vars or credentials and only needs file-read access to the target being scanned and write access to an output dir — which is proportional. Caveat: optional --upload-url will transmit a JSON report (instance-id required) to an external endpoint; while authors claim no source code or credentials are sent, scan findings and metadata could still contain sensitive information. Confirm the upload endpoint and the exact payload before enabling uploads.

Persistence & Privilege

The skill does not request always:true, does not require persistent presence, and does not declare changes to other skills or system-wide settings. Autonomous invocation is allowed by default (platform normal), but that alone is not a red flag here.

What to consider before installing

This skill claims to be a local scanner and documents sensible defaults, but there are two things to check before running it: (1) the SKILL.md/INSTALL.md refer to a bundled binary (assets/bin/skillscan) and provide a SHA-256, yet the package manifest you were shown does not include that binary—do not run or trust any binary unless you can locate it and verify the checksum; (2) the optional --upload-url and --engine external flags can cause reports or engine work to run/send data externally—only enable them if you trust the endpoint and understand exactly what the report contains. Preferred safe path: clone the listed repository and build from source, manually review the code (or have someone you trust do so), and run the scanner in a sandboxed environment.

Skill Scan Wrapper

当你要在安装一个本地 skill、归档或 release bundle 前做一次快速安全检查时，使用这个 skill。

⚠️ Security Notice

This tool operates locally and requires user trust in the binary you run. Always verify the checksum after downloading. For maximum security, build from source (recommended).

Binary Included

Property	Value
Location	assets/bin/skillscan
Version	v0.8.0
Platform	Linux ARM64
SHA-256	ee7fd87a3ad72984fcd60ba3adae1020fe7099d24332b7cc30e66034cd745dd7

Verify locally before running:

sha256sum assets/bin/skillscan
# Compare output with the SHA-256 value above

This bundled package includes a pre-compiled binary. You can still build from source if you prefer:

git clone https://gitee.com/random_player/cmic-skill-scanner.git
cd cmic-skill-scanner && cargo build --release

前置条件

• 默认不需要任何外部依赖
• --upload-url 和 --engine external 功能默认禁用，仅在用户显式配置时启用

信任模型

This is an open-source (MIT-0) package. The binary (bundled or downloaded) is a convenience only — it does not grant any additional trust.

Your options:

Approach	Trust Requirement	Verification
Build from source	None (you control everything)	Manual code review
Bundled/downloaded binary	You trust the release host	SHA-256 checksum

What the tool does NOT do by default:

• Does NOT upload data anywhere
• Does NOT connect to the network
• Does NOT access credentials, SSH configs, or environment variables
• Does NOT execute external tools unless you explicitly configure --engine external

工作流程

1. 调用 skillscan：

skillscan review /path/to/target --format markdown
skillscan review /path/to/skills --output-dir /tmp/skillscan-out

2. 阅读输出中的：输入类型、完整度、engine 执行状态、findings

网络上传功能 (默认禁用)

⚠️ This feature is completely optional and disabled by default. It requires explicit user configuration via --upload-url.

What gets sent (only when you configure --upload-url):

• A structured JSON report containing detection findings
• An instance identifier you supply via --instance-id
• No skill source code, credentials, or system configuration is ever transmitted

外部引擎集成 (默认禁用)

⚠️ This feature is completely optional and disabled by default. It requires explicit user configuration via --engine external.

Delegates pattern-matching to a user-configured local tool. This runs locally — no remote calls are made.

Permissions Required

Scope	Reason
Read files in target path	To analyze skill source code for patterns
Write to --output-dir	To save scan reports locally
Execute binary	To run the scanner engine
Network (optional)	Only if --upload-url is explicitly configured