Sharpen — Git skill for Claude Code
Cut hedges and weasel words; commit to claims without inflating them.
How to install Sharpen
Installs to ~/.claude/skills/marihq-mari-cc-sharpen/SKILL.md
mkdir -p ~/.claude/skills/marihq-mari-cc-sharpen && curl -fsSL https://raw.githubusercontent.com/MariHQ/mari-cc/HEAD/commands/sharpen.md -o ~/.claude/skills/marihq-mari-cc-sharpen/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Sharpen does
description: Cut hedges and weasel words; commit to claims without inflating them argument-hint: "" allowed-tools: Bash(mari *), Read, Edit
Run the **sharpen** editorial verb from the mari skill on $ARGUMENTS (a path, a natural-language reference like "the changelog", or — if omitted — the file(s) just edited this session; ask if none).
Setup first: load PRODUCT.md / STYLE.md / FACTS.md if present, read a representative file for voice, resolve the register, and run `mari detect
Alternatives in Git
- Worktree Deliver — Commit all work, push, and create a pull request from the current worktree 23.4k ★
- OSS PR — Commit + PR workflow for open source contributions: auto-creates a user-prefixed branch when on main/master, t 19.7k ★
- Yeet — Stage, commit, push code, and open a GitHub pull request via CLI 14.6k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Ship Improvements
Implement every open row in IMPROVEMENTS.md, close them, update the docs, verify, commit, push, wait for CI gr
Lane Safety
Safety and hygiene lane. Own redaction and every repository hygiene guard. Nothing derived from a game log rea
Nap
Mid-session mini-consolidation — brief reflection + optional commit without the full /sleep overhead
Automated Sde Skill
Claude Code skills for the SDE loop — plan, review, PR, resolve — automated end to end: they commit, push, ope
Verify PR
Verify that PR claims match actual implementation
Verify Claims
Verify factual claims in documents against code repository using git evidence
Related Agents
Evaluate
Evidence-based auditor in a fresh context. Verifies every claim in one named artifact (a spec cut, a board's c
Fable Advisor
Read-only second-opinion advisor for a decision at a commitment boundary, pinned to the roster's strongest-jud
Completion Gate
Lie detector for agent claims — validates every completion claim has actual evidence. Default-FAIL mindset. Us