gpt
- Repo stars 330
- License MIT
- Author updated Live
- Author repo claude-skill-registry
- Domain
- Data
- Compatible agents
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- Trust score
- 94 / 100 · audit passed
- Author / version / license
- @majiayu000 · MIT
- Token usage
- Lean
- Setup complexity
- Plug-and-play
- External API key
- Not required
- Operating systems
- Unspecified (assume cross-platform)
- Runtime requirements
- No special requirements
- Permissions
-
- Read-only
- Write / modify
- Network behavior
- Local-only
- Install commands
- 26 variants
Profile is derived at build time from SKILL.md and install vectors. Subject to drift from author intent.
Heads up: 未限定 allowed-tools，默认拥有全部工具权限。
---
name: gpt
description: Use GPT-5.2 for long-running coding tasks: large refactors, feature implementation etc. Long-con…
category: data
runtime: no special runtime
---
# gpt output preview
## PART A: Task fit
- Use case: Use GPT-5.2 for long-running coding tasks: large refactors, feature implementation etc. Long-context coding. Best for: large refactors, feature implementation. GPT strictly follows instructions. Provide as much high-level design context as possible: runs entirely locally. Works with Claude Code, Cursor, Cline and 23 more..
- Inputs: target material, constraints, expected output, and acceptance criteria.
- Evidence boundary: follow “Prompt Guidelines / Create Session (with worktree for code edits) / Create Session (read-only, no worktree)” and do not present inference as author intent.
## PART B: Execution result
- **01** The card summarizes the use case; runtime output centers on “Use GPT-5.2 for long-running coding tasks: large refactors, feature implementation etc. Long-context coding. Best for: large refactors, feature implementation. GPT strictly follows instructions. Provide as much high-level design context as possible: runs entirely locally. Works with Claude Code, Cursor, Cline and 23 more.”.
- **02** When the source has headings, the agent prioritizes “Prompt Guidelines / Create Session (with worktree for code edits) / Create Session (read-only, no worktree)” so the result follows the author’s structure.
- **03** Typical output includes task judgment, concrete steps, required commands or file edits, validation, and follow-up options.
- **04** Risk context follows the fingerprint: read files, write/modify files; mostly runs locally; usually needs no extra API key.
## Running Rules
- read files, write/modify files; mostly runs locally; usually needs no extra API key.
- Validate with a small sample before expanding scope.
- Return the result, validation criteria, and next iteration options. Choose the trigger
The source mentions slash commands such as `/path`; use them first when your agent supports command triggers.
Provide inputs and boundaries
Name target files or source material, expected output, forbidden changes, and whether network or shell access is allowed. Permission fingerprint: read files, write/modify files.
Validate with a small run
Start with a small task and check whether the result follows “Prompt Guidelines / Create Session (with worktree for code edits) / Create Session (read-only, no worktree)”. Inspect diffs, logs, previews, or tests before expanding scope.
Review before delivery
Confirm the final output includes a concrete result, evidence, and next action. If it stays generic, tighten inputs, boundaries, and acceptance criteria.
---
name: gpt
description: Use GPT-5.2 for long-running coding tasks: large refactors, feature implementation etc. Long-con…
category: data
source: majiayu000/claude-skill-registry
---
# gpt
## When to use
- Use GPT-5.2 for long-running coding tasks: large refactors, feature implementation etc. Long-context coding. Best for:…
- Use it when the task has clear inputs, repeatable steps, and validation criteria.
## What to provide
- Target material, scope, expected result, and forbidden changes.
- Whether network, commands, file writes, or external services are allowed.
## Execution rules
- Organize steps around “Prompt Guidelines / Create Session (with worktree for code edits) / Create Session (read-only, no worktree)” and keep inference separate from source facts.
- read files, write/modify files; mostly runs locally; usually needs no extra API key.
- Validate with a small sample before expanding the task.
## Output requirements
- Return the deliverable, key evidence, validation method, and next action.
- Mark missing information as unknown; do not invent commands, platforms, or dependencies. Evidence boundary and execution chain
The author source anchors workflow facts; repository files anchor sources and commands; Fluxly only adds fit, limitations, and quality judgment.
skill "gpt" {
input -> user goal + target files + boundaries + acceptance criteria
context -> Prompt Guidelines / Create Session (with worktree for code edits) / Create Session (read-only, no worktree)
rules -> SKILL.md triggers / order / output contract
runtime -> no special runtime | read files, write/modify files | mostly runs locally
guardrails -> usually needs no extra API key + small-sample validation + diff/log review
output -> copyable result + checklist + next iteration
}GPT-5.2 (via ocw)
Long-context coding. Best for: large refactors, feature implementation.
Prompt Guidelines
GPT strictly follows instructions. Provide as much high-level design context as possible:
- Architecture decisions, data flow, component responsibilities
- Constraints, edge cases, expected behaviors
Avoid:
- Contradictory requirements (GPT will struggle to reconcile conflicts)
- Code snippets — GPT writes code well on its own; use tokens for design info instead
Create Session (with worktree for code edits)
ocw new gpt --worktreeReturns:
- Line 1: 6-char hash
- Line 2: worktree path (e.g.,
/path/to/ocw-abc123)
The worktree is an isolated git branch. Work there freely.
Create Session (read-only, no worktree)
ocw new gptChat
ocw chat <hash> << 'EOF'
your prompt
EOFChat with File
ocw chat <hash> -f /path/to/spec.md << 'EOF'
implement based on this
EOFWorktree Workflow
ocw new gpt --worktree→ get hash + path- Work in worktree:
cd /path/to/ocw-{hash} - Edit, commit, push as needed
- When done:
git worktree remove /path/to/ocw-{hash}
List Sessions
ocw list
Decide Fit First
Design Intent
How To Use It
Boundaries And Review