thread-branching/SKILL.md
Create a ready-to-paste prompt for starting one or more new conversations from the current thread context. Use when Codex needs to hand off the current repo state, goal, constraints, touched files, verification status, or open risks into a self-contained prompt for another thread, subagent, or parallel follow-up conversation.
npx skillsauth add raphaeldelio/agent-skills thread-branchingInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill when the user wants to continue the current work in a new thread without losing context.
The output should be a prompt that can be pasted into a fresh conversation and still make sense on its own.
Produce a ready-to-paste prompt, not a plan about the prompt.
Unless the user asks otherwise:
Every handoff prompt should include:
Prefer concrete facts over summaries like "we changed some styles".
Decide what the new thread is for:
If the user wants multiple parallel threads, split them by file ownership or concern so they do not overlap.
Pull in only the current facts the next thread needs:
Do not dump the whole conversation.
The prompt should tell the next thread:
Use the templates in references/templates.md.
The new thread should not depend on hidden history.
Replace vague references like:
with explicit statements:
development
Plan and operate project-agnostic parallel feature work using atomic file-owned packets, dependency waves, handoff prompts, and repair packets. Use when Codex needs to split any feature, migration, refactor, or platform change into independent subagent tasks, define packet contracts, decide which packets can run in parallel, generate handoff prompts, review completed packets, or create narrow follow-up repair packets.
development
Maintainer-only workflow for handling GitHub Secret Scanning alerts on OpenClaw. Use when Codex needs to triage, redact, clean up, and resolve secret leakage found in issue comments, issue bodies, PR comments, or other GitHub content.
development
Maintainer workflow for OpenClaw releases, prereleases, changelog release notes, and publish validation. Use when Codex needs to prepare or verify stable or beta release steps, align version naming, assemble release notes, check release auth requirements, or validate publish-time commands and artifacts.
development
Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.