Agent skill
conversation-fork
Fork the current conversation to safely explore an alternative approach without losing current progress
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/conversation-fork
SKILL.md
Conversation Fork
Create a safe divergence point to explore an alternative approach without losing current progress.
When to Run
- "I'm not sure whether to do X or Y — let me try both"
- Before a risky refactor that might not pan out
- When the user wants to explore an alternative design
- After a plan is agreed but before major implementation begins
Arguments
- First argument: fork name (short, kebab-case, e.g.
auth-approach-b) --worktree: use heavy fork (new git worktree + branch) instead of light fork (git stash)
Strategy Selection
| Strategy | When | Cost |
|---|---|---|
| Light fork (default) | Experimenting, < 1 day of work, same branch OK | Low |
Heavy fork (--worktree) |
Long-lived parallel track, needs own CI, separate PR | High |
Light Fork (Default)
Creates a stash snapshot and documents both approaches in a FORK file.
Steps
- Capture state:
FORK_NAME="<name>"
BRANCH=$(git branch --show-current)
SHA=$(git rev-parse --short HEAD)
DATE=$(date +%Y-%m-%d)
- Stash current work (preserves staged + unstaged + untracked):
git stash push -m "fork/$FORK_NAME" --include-untracked
- Write fork document at
docs/plans/FORK-${FORK_NAME}.md:
# Fork: <NAME>
**Created:** <DATE>
**Branch:** <BRANCH>
**Commit:** <SHA>
## Context
<One paragraph: why we forked here, what decision point this represents>
## Approach A (Stashed / Original Path)
**Status:** Stashed as `fork/<NAME>`
**What it does:** <Description of the current approach>
**Resume:** `git stash apply stash^{/fork/<NAME>}`
## Approach B (Current / Exploration)
**Status:** Working from <SHA>
**What it will do:** <Description of the alternative>
**Resume:** Continue working (this is the active branch)
## Recent Context
<output of: git log --oneline -5>
## Decision Criteria
- Choose A if: <condition>
- Choose B if: <condition>
- Merge both if: <condition>
## Resume Instructions
To return to Approach A:
\`\`\`bash
git stash apply stash^{/fork/<NAME>}
\`\`\`
To abandon Approach B and restore A:
\`\`\`bash
git checkout -- .
git stash pop
\`\`\`
Heavy Fork (--worktree)
Creates a full git worktree at a new branch for true parallel development.
Steps
-
Capture state (same as light fork).
-
Create worktree at current HEAD:
WORKTREE_PATH="$HOME/worktrees/property-tracker/fork-${FORK_NAME}"
git worktree add "$WORKTREE_PATH" -b "fork/$FORK_NAME"
- Copy env:
cp .env.local "$WORKTREE_PATH/.env.local" 2>/dev/null || true
- Write fork document to both worktrees (
docs/plans/FORK-${FORK_NAME}.md):
# Fork: <NAME>
**Created:** <DATE>
**Origin Branch:** <BRANCH>
**Origin Commit:** <SHA>
## Approach A (Origin)
**Path:** <original worktree path>
**Branch:** <BRANCH>
**Resume:** Work here, this worktree is untouched
## Approach B (Exploration)
**Path:** <WORKTREE_PATH>
**Branch:** `fork/<NAME>`
**Resume:** `cd <WORKTREE_PATH>`
## Recent Context
<output of: git log --oneline -5>
## Cleanup
When done exploring, remove the worktree:
\`\`\`bash
git worktree remove <WORKTREE_PATH>
git branch -d fork/<NAME>
\`\`\`
Or keep it and open a separate PR from `fork/<NAME>`.
After Forking
- Light fork: You're on a clean working tree at the same SHA. Implement Approach B. Return to A with
git stash apply. - Heavy fork: Switch to the new worktree path. Both tracks run independently.
In either case, docs/plans/FORK-<name>.md records the decision context for future reference.
Examples
/conversation-fork auth-approach-b
/conversation-fork db-indexing-strategy --worktree
/conversation-fork no-orm-experiment --worktree
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
majiayu000/claude-skill-registry
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
majiayu000/claude-skill-registry
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
majiayu000/claude-skill-registry
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
majiayu000/claude-skill-registry
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
majiayu000/claude-skill-registry
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
majiayu000/claude-skill-registry
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?