explaingit

thananon/9arm-skills

Analysis updated 2026-06-24

1,133ShellAudience · developerComplexity · 1/5Setup · easy

TLDR

Personal collection of Claude Code agent skills sorted into engineering, productivity, misc, personal, and in-progress buckets. Ships a link-skills shell script that symlinks each into ~/.claude/skills/.

Mindmap

mindmap
  root((9arm-skills))
    Inputs
      SKILL.md files
      link-skills.sh
      Claude Code
    Outputs
      Symlinked skills
      Auto-invoked routines
      Structured outputs
    Use Cases
      Debugging discipline
      Post-mortems
      PR reviews
      Management writing
    Tech Stack
      Shell
      Markdown
      Claude Code
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Install the debug-mantra skill so Claude Code enforces a reproduce-trace-falsify-cross-check routine before any fix.

USE CASE 2

Use the post-mortem skill to write a structured engineering record after fixing a tricky bug.

USE CASE 3

Run the scrutinize skill on a draft pull request to get an outsider-style review.

USE CASE 4

Rewrite engineer-language updates into JIRA, Slack, or email tone with the management-talk skill.

What is it built with?

ShellMarkdownClaude Code

How does it compare?

thananon/9arm-skillswangnov/codex-app-mirrorhypnguyen1209/offensive-claude
Stars1,1331,256106
LanguageShellShellShell
Last pushed2026-07-02
MaintenanceActive
Setup difficultyeasyeasyeasy
Complexity1/51/53/5
Audiencedeveloperdeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 5min

Just clone and run link-skills.sh. The script assumes Claude Code is installed and reads from ~/.claude/skills/.

License is not stated in the available content.

In plain English

This repository is a personal collection of agent skills written for Claude Code, which is the command line coding assistant from Anthropic. A skill in this setting is a small folder that tells the assistant how to perform a specific kind of task. Each skill has a file called SKILL.md with a short header that names it and describes when it should be used, and the assistant pulls the skill in automatically when the situation matches. The README says the skills live in a folder named skills and are sorted into buckets. The engineering bucket holds daily code work, productivity holds daily non-code workflow tools, misc holds things that are kept around but rarely used, personal holds entries tied to the author's own setup that are not promoted, in-progress holds drafts that are not yet ready to ship, and deprecated holds skills no longer in use. Installation is simple. There is a shell script called link-skills.sh that creates symbolic links from every shippable skill into the user's ~/.claude/skills/ folder, which is where Claude Code looks for them. A second script called list-skills.sh prints every SKILL.md file in the repository so the author can see what is available. The README lists the skills that are currently ready. Under engineering there are three. The debug-mantra skill enforces a four-step debugging routine that reproduces the bug, traces the failure path, tries to falsify the working hypothesis, and cross-references every breadcrumb before any fix is attempted. The post-mortem skill writes a structured engineering record of a fixed bug, and it refuses to draft one unless there is a reliable reproduction, a known cause, and a validated fix. The scrutinize skill performs an outsider-style end-to-end review of a plan or pull request, asking whether the change does what it claims and whether there is a simpler way. Under productivity there is one skill. The management-talk skill rewrites engineer-to-engineer content for an engineering management audience, and it adjusts the wording for the specific channel, such as a JIRA ticket, a Slack message, an async standup, an email, or a set of meeting talking points. The misc bucket is empty for now.

Copy-paste prompts

Prompt 1
Clone thananon/9arm-skills, run link-skills.sh, then verify all engineering skills are linked into ~/.claude/skills with a single ls command.
Prompt 2
Open skills/engineering/debug-mantra/SKILL.md and explain the four-step routine. Then apply it to this failing test: <paste log here>.
Prompt 3
Use the scrutinize skill from this repo to review my draft PR. Here is the diff: <paste here>. Focus on whether the change does what it claims.
Prompt 4
Fork 9arm-skills, add a new skill under engineering called pair-debug, and write its SKILL.md header so Claude Code auto-invokes it on Python tracebacks.
Prompt 5
Compare the post-mortem skill format here with the Google SRE post-mortem template. List every section the SRE template has that this skill omits.

Frequently asked questions

What is 9arm-skills?

Personal collection of Claude Code agent skills sorted into engineering, productivity, misc, personal, and in-progress buckets. Ships a link-skills shell script that symlinks each into ~/.claude/skills/.

What language is 9arm-skills written in?

Mainly Shell. The stack also includes Shell, Markdown, Claude Code.

What license does 9arm-skills use?

License is not stated in the available content.

How hard is 9arm-skills to set up?

Setup difficulty is rated easy, with roughly 5min to a first successful run.

Who is 9arm-skills for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.