parse-git-status
Parse git status output into structured data showing staged, modified, and untracked files. Use for pre-flight validation, checking clean working directory, or listing changed files before commits.
When & Why to Use This Skill
This Claude skill transforms raw Git status output into structured JSON data, facilitating programmatic analysis of repository states. It enables seamless tracking of staged, modified, and untracked files, making it an essential utility for automating version control workflows, enhancing developer productivity, and ensuring repository health through machine-readable insights.
Use Cases
- Automated Pre-flight Checks: Use structured data to verify a clean working directory or ensure specific files are staged before executing deployment scripts or automated commits.
- CI/CD Pipeline Integration: Integrate parsed git metadata into local automation scripts to conditionally trigger builds or tests based on the specific files or directories modified.
- AI-Assisted Development: Provide Claude agents with a clear, structured view of local changes to help generate accurate, context-aware commit messages or pull request descriptions.
- Repository Health Auditing: Programmatically scan local environments to identify uncommitted changes, detached HEAD states, or branches that are ahead/behind their remote counterparts.
| name | parse-git-status |
|---|---|
| description | Parse git status output into structured data showing staged, modified, and untracked files. Use for pre-flight validation, checking clean working directory, or listing changed files before commits. |
Parse Git Status
Parse git status command output into structured JSON for programmatic analysis.
When to Use
- Pre-flight checks before workflow execution
- Validate clean working directory
- List modified files for commit
- Check for uncommitted changes
Instructions
Step 1: Receive Git Status Output
Accept raw git status output as input.
Expected Input:
gitStatusOutput: String (raw output fromgit status --porcelainor regulargit status)
Step 2: Parse Branch Information
Extract current branch and tracking information.
Patterns:
## branch-name: Current branch## branch-name...origin/branch-name: Tracking branch[ahead N]or[behind N]: Ahead/behind commits
Step 3: Categorize Files
Parse file status indicators and categorize.
Status Indicators (porcelain format):
M: Modified (staged)M: Modified (unstaged)A: Added (staged)D: Deleted (staged)R: Renamed (staged)??: Untracked!!: Ignored
Step 4: Return Structured Data
Return parsed data as JSON object.
Expected Output:
{
"branch": "main",
"tracking": "origin/main",
"ahead": 0,
"behind": 0,
"staged": ["file1.ts", "file2.ts"],
"modified": ["file3.ts"],
"deleted": [],
"renamed": [],
"untracked": ["file4.ts"],
"clean": false
}
Error Handling
- Invalid Git Output: Return error describing format issue
- Not a Git Repository: Return error indicating no git repo
- Empty Output: Return clean status with empty arrays
Examples
Example 1: Clean Working Directory
Input:
On branch main
Your branch is up to date with 'origin/main'.
nothing to commit, working tree clean
Output:
{
"branch": "main",
"tracking": "origin/main",
"ahead": 0,
"behind": 0,
"staged": [],
"modified": [],
"deleted": [],
"renamed": [],
"untracked": [],
"clean": true
}
Example 2: Modified Files
Input (porcelain format):
## main...origin/main [ahead 2]
M src/utils.ts
M src/types.ts
A src/new-feature.ts
?? temp-file.js
Output:
{
"branch": "main",
"tracking": "origin/main",
"ahead": 2,
"behind": 0,
"staged": ["src/utils.ts", "src/new-feature.ts"],
"modified": ["src/types.ts"],
"deleted": [],
"renamed": [],
"untracked": ["temp-file.js"],
"clean": false
}
Example 3: Detached HEAD
Input:
## HEAD (no branch)
M README.md
Output:
{
"branch": "HEAD (detached)",
"tracking": null,
"ahead": 0,
"behind": 0,
"staged": [],
"modified": ["README.md"],
"deleted": [],
"renamed": [],
"untracked": [],
"clean": false
}
Validation
- Parses branch information correctly
- Categorizes files by status
- Handles empty/clean status
- Parses ahead/behind indicators
- Handles detached HEAD state
- Returns clean:true only when appropriate
Supporting Files
None required - pure parsing logic.