How do I install the Whats Next Mode skill?

Run npx skillmds@latest add jordancoin/whats-next-mode in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Whats Next Mode safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Whats Next Mode?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Whats Next Mode free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Whats Next Mode?

jordancoin (@jordancoin) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Whats Next Mode by jordancoin — Agent Skill

Name: Whats Next Mode
Author: jordancoin

Whats Next Mode

Switch this machine's whats-next-mcp wiring between DEV (local ./dist build) and WILD (published global npm bins), toggling both the Claude Code Stop hook and the whats-next MCP server in one shot. Use when the user says "switch to dev/wild/local/prod whats-next", "use the local build", "use the published version", "what mode am I in", or wants to toggle between developing whats-next-mcp and using the released package.

jordancoin Updated Sep 17, 2026

File contents

whats-next-mode

Toggle whats-next-mcp between two setups, both the Stop hook and the MCP server at once, via scripts/switch_mode.py.

DEV — hook + MCP run node <repo>/dist/..., so they reflect your latest npm run build. Use while developing whats-next-mcp itself.
WILD — hook + MCP run the global whats-next-* bins from npm install -g. Repo-independent; the released version real users get.

How to run it

Always start with status to see where things stand, then switch. Prefer --dry-run first when the user is unsure.

python3 scripts/switch_mode.py status          # show current wiring
python3 scripts/switch_mode.py dev --build      # local build (rebuild first)
python3 scripts/switch_mode.py wild             # published global bins
python3 scripts/switch_mode.py dev --dry-run    # preview, change nothing

Run the command from the repo root. Report the output back to the user.

Flags

Flag	Effect
`dev` / `wild` / `status`	mode (aliases: `local`→dev, `prod`/`npm`/`global`→wild)
`--build`	run `npm run build` first (dev only) so dist is fresh
`--gstack` / `--no-gstack`	gstack-skill picker vs generic (default: gstack on)
`--scope user\|local\|project`	MCP registration scope (default: local)
`--settings <path>`	which settings.json to edit (default: project `.claude/settings.json`)
`--no-hook` / `--no-mcp`	only touch one of the two
`--dry-run`	print the changes without applying them

After switching — IMPORTANT

The script edits .claude/settings.json and re-registers the MCP server, but Claude Code caches both at session start. Tell the user to run /hooks (or restart) to reload. Until they do, the running session keeps the old wiring.

Notes

dev mode reflects local code edits after each npm run build; wild mode only changes when you npm publish + npm update -g.
The script never clobbers a malformed settings.json — it skips the hook edit and says so.
It's idempotent: running the same mode twice is a no-op beyond rewriting the command string.

---name: whats-next-modedescription: Switch this machine's whats-next-mcp wiring between DEV (local ./dist build) and WILD (published global npm bins), toggling both the Claude Code Stop hook and the whats-next MCP server in one shot. Use when the user says "switch to dev/wild/local/prod whats-next", "use the local build", "use the published version", "what mode am I in", or wants to toggle between developing whats-next-mcp and using the released package.---# whats-next-modeToggle whats-next-mcp between two setups, both the **Stop hook** and the **MCPserver** at once, via `scripts/switch_mode.py`.- **DEV** — hook + MCP run `node <repo>/dist/...`, so they reflect your latest `npm run build`. Use while developing whats-next-mcp itself.- **WILD** — hook + MCP run the global `whats-next-*` bins from `npm install -g`. Repo-independent; the released version real users get.## How to run itAlways start with `status` to see where things stand, then switch. Prefer`--dry-run` first when the user is unsure.```bashpython3 scripts/switch_mode.py status # show current wiringpython3 scripts/switch_mode.py dev --build # local build (rebuild first)python3 scripts/switch_mode.py wild # published global binspython3 scripts/switch_mode.py dev --dry-run # preview, change nothing```Run the command from the repo root. Report the output back to the user.## Flags| Flag | Effect ||------|--------|| `dev` / `wild` / `status` | mode (aliases: `local`→dev, `prod`/`npm`/`global`→wild) || `--build` | run `npm run build` first (dev only) so dist is fresh || `--gstack` / `--no-gstack` | gstack-skill picker vs generic (default: gstack on) || `--scope user\|local\|project` | MCP registration scope (default: local) || `--settings <path>` | which settings.json to edit (default: project `.claude/settings.json`) || `--no-hook` / `--no-mcp` | only touch one of the two || `--dry-run` | print the changes without applying them |## After switching — IMPORTANTThe script edits `.claude/settings.json` and re-registers the MCP server, butClaude Code caches both at session start. **Tell the user to run `/hooks` (orrestart) to reload.** Until they do, the running session keeps the old wiring.## Notes- `dev` mode reflects local code edits after each `npm run build`; `wild` mode only changes when you `npm publish` + `npm update -g`.- The script never clobbers a malformed `settings.json` — it skips the hook edit and says so.- It's idempotent: running the same mode twice is a no-op beyond rewriting the command string.

Frequently asked questions