AgentStack
SKILL verified MIT Self-run

Commit

skill-giladresisi-ai-dev-env-commit · by giladresisi

Use when the user asks to commit changes, create a commit, or save work to git

No reviews yet
0 installs
13 views
0.0% view→install

Install

$ agentstack add skill-giladresisi-ai-dev-env-commit

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No 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 Used
  • 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.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README — it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-giladresisi-ai-dev-env-commit)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.

Preview Execution monitoring

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 →
Are you the author of Commit? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Create a new commit for all of our uncommitted changes

Step 0: Determine Reasoning

Before committing, identify WHY these changes were needed. The commit body MUST include a short reason.

  • If the user explicitly stated a reason (e.g., "commit my changes, I fixed the login bug"), use that.
  • If this skill was invoked by another skill/command and the reason is clear from context, use that.
  • If the reason is NOT clear from the conversation context, use AskUserQuestion to ask: "Why were these changes needed? (This will be included in the commit message)"

Step 1: Review Current State

Run: git status && git diff HEAD && git status --porcelain This shows all uncommitted changes (modified, untracked, and staged files)

Step 1.5: Secret / Credential Scan

Before staging anything, scan the diff output from Step 1 for explicit secrets or credentials.

Two complementary checks — BOTH must pass:

Check A: Value-pattern scan

Look for values that look like secrets regardless of what field they are in:

  • API keys with known prefixes (e.g., sk-..., AIza..., AKIA..., ghp_..., xoxb-...)
  • Private keys or certificates (e.g., -----BEGIN RSA PRIVATE KEY-----)
  • Hard-coded connection strings with credentials (e.g., postgresql://user:pass@host)
  • Any high-entropy string (20+ random chars) assigned to a variable named key, secret, token, password, credential, or similar

Check B: Field-name scan (catches low-entropy keys like org-agentic-kb)

Scan for any non-placeholder value assigned to an authentication-related field name, regardless of the value's format or length. Flag any of these patterns:

"X-API-Key": ""
"Authorization": "..."
api_key: ...
apiKey: ...
token: ...
password: ...
secret: ...
credential: ...

In JSON, YAML, .env, config files, or any headers block (e.g., MCP mcp.json, CI configs).

A value is safe (not flagged) only if it is clearly a placeholder:

  • Wrapped in angle brackets: ``
  • Prefixed with PLACEHOLDER_, YOUR_, MY_, EXAMPLE_
  • An empty string: ""
  • A well-known dummy: "changeme", "test-key" (flag these too — they suggest a real value is expected)

If any suspected secret is found:

  1. DO NOT stage or commit anything.
  2. Alert the user with the exact file path, line number, and the suspicious value (mask the middle characters, e.g., org-ag...kb).
  3. Ask the user to confirm whether it is a real secret and how to proceed before continuing.
  4. STOP — do not proceed to Step 2 until the user responds.

Only continue to Step 2 if no secrets were found, or the user has confirmed the flagged values are safe.

Step 2: Stage ALL Changes

CRITICAL: You MUST run git add . to stage ALL files including:

  • Modified files
  • Untracked files
  • Untracked directories

DO NOT selectively stage files with git add unless explicitly instructed. The default behavior is to commit everything.

Run: git add .

Step 3: Create Commit

Add an atomic commit message with conventional commit format:

  • Type: "feat", "fix", "docs", "chore", "refactor", "test", etc.
  • Scope (optional): component or area affected
  • Description: brief summary (50 chars or less, imperative mood)
  • Body (REQUIRED): must include a short reason explaining WHY these changes were needed
  • Footer: Include "Co-Authored-By: Claude Sonnet 4.5 "

Run: `git commit -m "$(cat

Co-Authored-By: Claude Sonnet 4.5 EOF )"`

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.