← Back to Skills Marketplace
73
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install nm-sanctum-commit-messages
Description
Generate conventional commit messages from staged changes with correct type/scope
Usage Guidance
This skill appears coherent and limited to generating commit messages from staged changes. Before using it, confirm you are comfortable letting the agent read your staged diffs and recent commit history (these may contain sensitive data). The skill writes a local file ./commit_msg.txt — review that file before running git commit. If you enable the optional 'sem' integration, ensure the sem binary is from a trusted source. As a best practice, run the skill in a trusted repo and inspect the generated message before committing.
Capability Analysis
Type: OpenClaw Skill
Name: nm-sanctum-commit-messages
Version: 1.0.0
The skill is designed to generate conventional commit messages by analyzing staged changes using standard git commands (git status, git diff, git log) and an optional semantic diff tool. It includes stylistic guidelines to avoid corporate jargon and follows standard git practices, such as ensuring pre-commit hooks are respected. No malicious behavior, data exfiltration, or unauthorized execution patterns were found in SKILL.md or _meta.json.
Capability Assessment
Purpose & Capability
Name/description claim to generate conventional commit messages from staged changes. The SKILL.md only references git commands, optional use of a 'sem' tool if present, and writing a local preview file — all directly related to that purpose.
Instruction Scope
Instructions tell the agent to run git status/diff/log on the repository and to write ./commit_msg.txt. This is expected, but it means the agent will read repository contents (staged hunks and recent commits). The SKILL.md also optionally invokes an external tool 'sem' when available; that tool is not required and is clearly optional.
Install Mechanism
No install spec or code files — instruction-only skill. Nothing is downloaded or written to disk by an installer. Lowest-risk install posture.
Credentials
No environment variables, credentials, or config paths are requested. The only required privileges are read access to the local Git repository and permission to create ./commit_msg.txt, which are proportional to the stated task.
Persistence & Privilege
always:false and default invocation settings. The skill writes a single local preview file and does not request persistent system-wide changes or modify other skills. No elevated privileges requested.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install nm-sanctum-commit-messages - After installation, invoke the skill by name or use
/nm-sanctum-commit-messages - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release.
- Generates conventional commit messages from staged git changes.
- Supports type/scope classification and entity-aware messages with sem integration.
- Enforces commit message formatting: subject, body, and footer with clear rules.
- Includes a "slop check" to avoid jargon and use plain language.
- Outputs message to commit_msg.txt and previews before commit.
Metadata
Frequently Asked Questions
What is Nm Sanctum Commit Messages?
Generate conventional commit messages from staged changes with correct type/scope. It is an AI Agent Skill for Claude Code / OpenClaw, with 73 downloads so far.
How do I install Nm Sanctum Commit Messages?
Run "/install nm-sanctum-commit-messages" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Nm Sanctum Commit Messages free?
Yes, Nm Sanctum Commit Messages is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Nm Sanctum Commit Messages support?
Nm Sanctum Commit Messages is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Nm Sanctum Commit Messages?
It is built and maintained by athola (@athola); the current version is v1.0.0.
More Skills