Rule Spec

Define, manage, and compile business rules as structured YAML data into LLM-ready prompts and agent-loadable SKILL.md files. Use when the user wants to creat...

This skill appears to do what it says, but exercise caution before running it. Key steps to reduce risk: - Verify the npm package: find the official rulespec package name, publisher, repository URL, and a release tag/commit. Do not run npx against an unknown package without inspection. - Prefer a pinned, audited install (add an explicit install spec or vendor the code) rather than 'npx' dynamic fetches. - Inspect the rulespec code (or its GitHub release) for any file I/O behavior, path handling, and network calls. - Run first in an isolated environment (container or VM) and with a non-privileged account. Back up your skills/ directory before using emit. - Avoid running this skill autonomously on an agent with wide filesystem privileges; restrict which paths it can read/write and validate any domain/outdir values to prevent overwrites or path traversal. - If you need to allow the CLI to read example files, limit inputs to non-sensitive test data and avoid passing system paths (e.g., do not pass /etc, home, or application config files). If you can obtain the package repository or a signed release, re-run this evaluation with that information — it would raise confidence and likely move the verdict toward benign if the source is legitimate and reviewed.

Type: OpenClaw Skill Name: rulespec Version: 1.0.7 The skill bundle provides a CLI-based workflow for managing business rules using the `rulespec` package via `npx`. It includes high-risk capabilities such as reading arbitrary files (via the `--input` flag in `add-example`) and writing to arbitrary directories (via the `--outdir` flag in `emit`), which could be exploited for data exfiltration or unauthorized file system modification. While these features are aligned with the stated purpose of rule management and compilation, the lack of path sanitization and the reliance on external code execution via `npx` constitute significant security risks in an agentic context.