agent-loop-ext
Crash-resilient external agent loop with state persistence and CI/CD integration
Disable the AIWG context hook without uninstalling AIWG or deleting hook files
$ npx -y skills add jmagly/aiwg --skill hook-disable --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/hook-disableContext preview
The summary Claude sees to decide when to auto-load this skill.
Disable the AIWG context hook without uninstalling AIWG or deleting hook files
namespace: aiwg name: hook-disable platforms: [all] description: Disable the AIWG context hook without uninstalling AIWG or deleting hook files commandHint: argumentHint: "[--provider claude|warp|copilot|cursor|factory|windsurf|opencode|codex] [--all]" allowedTools: Read, Write, Edit, Bash, Glob model: haiku category: hook-management modelRole: efficiency modelTier: economy
You are an AIWG Hook Management Specialist responsible for temporarily disabling the AIWG context hook.
Remove the AIWG hook directive from the platform context file(s) so AIWG context is NOT loaded at the next session start. The hook file (`AIWG.md` or equivalent) is **preserved on disk** — no content is lost.
| Flag | Description | |------|-------------| | `--provider <name>` | Target specific provider: `claude`, `warp`, `copilot`, `cursor`, `factory`, `windsurf`, `opencode`, `codex` | | `--all` | Disable for all installed providers (default if no provider specified) |
| Provider | Context File | Directive to Remove | |----------|-------------|---------------------| | Claude Code | `CLAUDE.md` | `@AIWG.md` | | Warp Terminal | `WARP.md` | `@AIWG-warp.md` or AIWG section | | Windsurf | `AGENTS.md` | `@AIWG-windsurf.md` | | GitHub Copilot | `.github/copilot-instructions.md` | `@AIWG-copilot.md` | | Cursor | `.cursorrules` | `@AIWG-cursor.md` | | Factory AI | `AGENTS.md` | `@AIWG-factory.md` | | OpenCode | `.opencode/context.md` | `@AIWG-opencode.md` | | Codex | `CODEX.md` | Full injection block (between AIWG markers) |
If `--provider <name>` specified, operate on that provider only. If `--all` or no flag, detect installed providers by checking for their context files.
For each target provider:
1. Check if the context file exists 2. Check if the hook directive is currently present
# Example for Claude Code grep -q "@AIWG.md" CLAUDE.md && echo "enabled" || echo "already disabled"
**For @-link style** (Claude, Warp, Windsurf, etc.):
Remove the `@AIWG-{provider}.md` line from the context file. Preserve all surrounding content.
**For Codex** (full injection):
Remove the block between AIWG injection markers:
<!-- BEGIN AIWG --> ...content... <!-- END AIWG -->
**For Warp** (section style):
Remove the `## AIWG Framework Context` section and its `@AIWG-warp.md` line, or comment it out.
After removing the directive, confirm:
# Hook file should still exist ls AIWG.md && echo "hook file preserved" || echo "WARNING: hook file missing"
Hook disabled for Claude Code Removed @AIWG.md directive from CLAUDE.md AIWG.md preserved at AIWG.md To re-enable: aiwg hook-enable To check status: aiwg hook-status
If the hook directive is already absent:
Hook already disabled for Claude Code No @AIWG.md directive found in CLAUDE.md No changes made.
| Condition | Action | |-----------|--------| | Context file missing | Report: nothing to disable | | Hook file missing | Report warning: hook disabled but hook file not found. Use `aiwg hook-regenerate` to recreate. | | No write permission | Report permission error |
`hook-disable` NEVER:
It only removes the single directive line that loads AIWG context.
# Disable for Claude Code only /hook-disable --provider claude # Disable for all installed providers /hook-disable --all # Disable (default: all installed) /hook-disable
Reusable project context and specialist workflows for the AI tools you already use. Plan software, coordinate specialist reviews, prepare campaigns, investigate incidents, organize research, curate media, and maintain operational knowledge.
Repo: jmagly/aiwg
Crash-resilient external agent loop with state persistence and CI/CD integration
Detect requests for iterative autonomous agent loops and route to the appropriate loop executor
Automatically execute tests when code-generating agents modify source files, enforcing the execute-before-return pattern
Enable agent loops to learn from similar past tasks and share patterns across loops
Query and manage the executable feedback debug memory
Execute tests on generated code and iterate until passing