Install
$ agentstack add skill-sjarmak-agent-workflows-simplify ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Review changed code for reuse, quality, and efficiency, then fix any issues found.
Arguments
$ARGUMENTS — format: [file path | glob pattern | blank]
- If a file path or glob is provided, scope the review to those files.
- If no argument is provided, review all uncommitted changes via
git diff.
Parse Arguments
Extract:
- scope: a file path, glob pattern, or empty (defaults to uncommitted changes)
Phase 1: Identify Changes
- If scope is empty, run
git diff --name-onlyto list changed files. If there are no uncommitted changes, checkgit diff --name-only HEAD~1for the last commit. - If scope is a path or pattern, resolve matching files.
- For each file, run
git diff(orgit diff HEAD~1if already committed) to get the actual changes. - Present the scope to the user: file count, total lines changed, and a one-line summary per file.
Phase 2: Analyze
For each changed file, look for:
- Unnecessary complexity — can this be simpler without losing clarity?
- Duplicated code — shared logic that should be extracted
- Dead code — unused imports, unreachable branches, commented-out code
- Over-engineering — abstractions that aren't needed yet
- Missing error handling — at system boundaries only (user input, external APIs)
Phase 3: Apply Fixes
Apply fixes directly — don't just report issues. For each fix:
- Make the change
- Verify the file still parses / compiles
- If tests exist, run them after all changes to confirm nothing broke
Rules
- Scope discipline: only refine recently modified code unless the user explicitly asks for broader review.
- Preserve functionality: never change what the code does, only how it does it.
- Prefer deletion over addition: the best code is code that doesn't exist.
- Clarity over brevity: explicit code is often better than compact code. Avoid:
- Overly clever solutions that are hard to understand
- Combining too many concerns into single functions
- Removing helpful abstractions that improve code organization
- Dense one-liners or nested ternaries that sacrifice readability
- Changes that make the code harder to debug or extend
- No cosmetic drift: don't reformat code you didn't change for functional reasons.
- Run tests: always verify after changes.
Pipeline Position
- Runs after implementation, before code review and commit.
- In
/focus, this is the simplify phase between execute and review. - Can be used standalone on any codebase at any time.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: sjarmak
- Source: sjarmak/agent-workflows
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.