Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Issuefinder Tool
v1.0.0功能强大的车辆日志下载与解析工具，支持云端日志下载、本地日志处理、自动故障分析等功能
⭐ 0· 346·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description (vehicle log download & analysis) align with the included scripts: wrapper + a larger CLI that downloads cloud logs, uploads files, runs tools, and extracts archives. However the client talks to a hardcoded internal-looking server (issuefinder-playground-init-dev.inner.chj.cloud) and performs remote environment creation/upload/download operations while the skill metadata declares no credentials or API tokens — unusual but not impossible for an internal unauthenticated service. This mismatch is noteworthy.
Instruction Scope
SKILL.md and WRAPPER_USAGE.md instruct the user to set an output path, alias a wrapper, and run the CLI. The runtime instructions do not tell the agent to read unrelated system files or scan the host; they only operate on user-specified log files or cloud downloads. The wrapper/CLI operate on file paths the user provides (expected for this tool).
Install Mechanism
There is no install spec (instruction-only) and the included scripts are run directly. No third-party downloads or archive extraction during install are declared. The tool itself performs network calls at runtime, but the repository does not install remote code during skill installation.
Credentials
The tool can upload arbitrary local log files and create remote environments via HTTP endpoints (/api/environment/create, /api/files/{env}/upload, etc.). The skill metadata requests no credentials or environment variables, yet the code transmits potentially sensitive logs to a remote host. There is no declaration/consent in SKILL.md about uploading logs to that server or what data is transmitted. This is a proportionality concern: either credentials/ownership should be declared and documented, or the documentation should explicitly warn users that files will be sent to an external API.
Persistence & Privilege
The skill is not always:true and is user-invocable (normal). The agent may invoke it autonomously (platform default). Autonomous invocation combined with the tool's ability to upload local files increases blast radius if the agent is permitted to call the skill without further checks. This is a contextual risk but not a direct misconfiguration in the skill metadata.
What to consider before installing
Before installing or running this skill: 1) Treat it as capable of sending local files to the remote server hardcoded in the code — do not run it on machines containing sensitive logs or PII. 2) Verify the remote endpoint (issuefinder-playground-init-dev.inner.chj.cloud) and whether it is trusted/within your organization; ask the author for hosting and auth details. 3) Prefer running the tool in an isolated/test environment and inspect or run the Python scripts locally to confirm what is uploaded. 4) If you need to allow network uploads, require explicit configuration of the target server and credentials (and a clear privacy/data-handling policy) rather than relying on a hardcoded endpoint. 5) If you do not want the agent to call this skill autonomously, restrict autonomous invocation in your agent settings or only invoke it manually.Like a lobster shell, security has layers — review code before you run it.
latestvk972j0mg6dq0tqw96rs6t92f4582450p
License
MIT-0
Free to use, modify, and redistribute. No attribution required.