next-skills

Surfaces declared successor skills with activation phrases after a participating skill completes.

14|1|Updated Jun 14, 2026
One-click install
npx skills add https://github.com/allemaar/open-skills --skill next-skills-allemaar
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: next-skills
Source: https://github.com/allemaar/open-skills/tree/main/skills/next-skills
Command: npx skills add https://github.com/allemaar/open-skills --skill next-skills-allemaar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After a skill finishes running, users often don't know what to do next. The Next Skills protocol (NSP) reads a completed skill's declared follow-on recommendations and presents them as a pickable list, so multi-skill workflows chain together without the user memorizing the catalog. ## Core Features & Use Cases - Successor surfacing: Reads the next-skills: front-matter list of a finished skill and offers each entry with its activation phrase and a one-line reason. - Runtime-mapped presentation: Uses AskUserQuestion multi-select in Claude Code, or a numbered prose list in Codex and other runtimes. - Offer-only, loop-safe design: Recommendations are never auto-invoked; the caller gates every hop, and empty lists stop silently. - Use Case: After an investigation skill completes, NSP offers /plan-create and /cold-review as next steps, letting the user pick the follow-on action with one selection. ## Quick Start Run /next-skills followed by a skill name to preview that skill's declared successor recommendations.

Frequently Asked Questions about next-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I recommend follow-up skills after a skill finishes?

Add a next-skills list to the skill's YAML front-matter with skill, phrase, and why fields for each successor. When the skill completes, the Next Skills protocol reads that list and offers the recommendations for the caller to pick.

How does the Next Skills protocol present recommendations in Claude Code?

In Claude Code it uses the AskUserQuestion tool with multi-select, one option per recommendation. In Codex or runtimes without a structured-question tool, it prints a numbered prose list and waits for the caller's reply.

Can next-skills auto-run the recommended skills?

No. Recommendations are offered only and never auto-invoked; the caller gates every hop. This prevents autonomous A-to-B-to-A loops, since each transition requires an explicit user choice.

What happens if a skill has no next-skills front-matter?

The protocol stops silently and surfaces nothing. If the next-skills skill itself is missing or unreadable, the opt-in block is a no-op and the participating skill completes normally.

What is the difference between next-skills and caller-options?

Caller-options runs before a skill executes to route its venue and mode, while next-skills runs after completion to recommend successor skills. A skill can carry both opt-in blocks; they operate independently.