Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Portal

ASecurity

Creates a git worktree for isolated feature work (and documents cleanup post-merge); respects the disableGit preference. Use to isolate a feature on its own branch/worktree.

14 stars
0 votes
0 copies
0 views
Added 9/28/2026
ai-agentsgobashgit

Works with

claude code

Security Analysis

A100/100

Scanned 9/28/2026

Install to Claude Code

$npx -y skills add Alexander-Tyagunov/magician --skill portal --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Portal?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Portal
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/alexander-tyagunov-portal/badge)](https://www.skillsdirectory.com/skills/alexander-tyagunov-portal)

More formats (shields.io, HTML) on the badges page.

Files
SKILL.md
---
name: portal
description: Creates a git worktree for isolated feature work (and documents cleanup post-merge); respects the disableGit preference. Use to isolate a feature on its own branch/worktree.
allowed-tools: Read, Bash(git worktree add *)
argument-hint: "[feature-name]"
---

# /portal — Git Worktree Isolation

Create an isolated git worktree for feature development.

## Check disableGit Mode

Read `.workspace/local/prefs.md` for `disableGit: true`. If set, skip all git operations and work in the current directory.

## Process (git mode)

1. **Get branch name** — if `$ARGUMENTS` is non-empty, use it as the feature name directly. Otherwise ask: "What's this feature called? (I'll use it as the branch/worktree name.)" **End your turn. Wait for their answer before creating anything.**
2. **Create worktree** (one plain command, so it matches this skill's pre-approval):
   ```bash
   git worktree add ../<repo-name>-<name> -b feature/<name>
   ```
3. **Workspace files** — tracked files (`.workspace/shared/`) come with the branch; gitignored ones such as `.workspace/local/prefs.md` don't. Worktrees that Claude Code creates itself (`claude --worktree`, subagent worktree isolation) copy the gitignored files listed in a `.worktreeinclude` file at the repo root, so listing `.workspace/local/prefs.md` there is a useful tip for the user. For this plain `git worktree add`, copy any local file the new worktree needs (that copy goes through the normal permission prompt).
4. **Confirm** the new worktree path to the user
5. Say: "Worktree created at `../<path>`. Work there for isolation. Run /seal when ready to merge."

## Process (disableGit mode)

1. Create a feature directory: `mkdir -p .features/<name>`
2. Note: no git isolation — be careful about conflicts with other in-progress work
3. Say: "Working in .features/<name>/ (disableGit mode — no worktree created)."

## Keep sibling worktrees in the loop

Worktrees isolate the files, not the consequences: a rename, a signature change, or a shared-dependency bump made here breaks whatever a sibling session is building on. When Claude Code's cross-session messaging is available, tell the affected session yourself instead of letting it discover the breakage — `ListAgents` finds the sessions working the other worktrees, `SendMessage` delivers one self-contained sentence about what landed.

Feature-detect it: no `ListAgents`, or no peer listed, means carry on exactly as before and note the change in your own summary. It isn't available in every environment, and a session inside a container can't see one on the host. See [lore/cross-session.md](../../lore/cross-session.md).

## Cleanup After Merge

After /seal completes and the PR merges:
```bash
git worktree remove ../<repo-name>-<name>
git branch -d feature/<name>
```

## Obstacles

This is a thin git-worktree utility. If a worktree cannot be created (branch already exists, dirty tree), report it in prose to the caller and stop; it emits no structured upward Obstacles block.

See [lore/obstacles.md](../../lore/obstacles.md).

## Completion Signal

"Portal open. Worktree at `<path>`, branch `<branch>`. Start implementation or run /orchestrate."

Attribution

Alexander-TyagunovAlexander-Tyagunov
View sourceMore from Alexander-Tyagunov →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

695601 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →