Skip to content
Agentmart
Submit a tool

vs-next

Use at the end of a piece of work or workflow phase, before selecting or starting another workflow, to decide whether to continue in the current session, use…

Skill by vltansky

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/vltansky-vs-next

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use at the end of a piece of work or workflow phase, before selecting or starting another workflow, to decide whether to continue in the current session, use a subagent, hand off, compact, clear, or stop. Trigger whenever the user asks what should happen now or next, whether to start fresh, or how to continue. This routing decision wins over a skill for future work when the user is asking how to proceed rather than asking to perform that future work.

Use cases: productivity, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/vs-next

What's inside

Skill name
vs-next
Description
Use at the end of a piece of work or workflow phase, before selecting or starting another workflow, to decide whether to continue in the current session, use a subagent, hand off, compact, clear, or stop. Trigger whenever the user asks what should happen now or next, whether to start fresh, or how to continue. This routing decision wins over a skill for future work when the user is asking how to proceed rather than asking to perform that future work.
License
MIT
Instructions
65 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (3)

  • SKILL.md
  • openai.yaml · agents/
  • vs-next.eval.ts · test/

SKILL.md (first lines)

# Next

Choose how work should cross the current phase boundary. The agent makes the
decision; the user should not need to remember the context-management model.

## Flow Contract

- **Kind:** Building block
- **Inputs:** Current objective, phase status, durable artifacts, expected next
  work, ownership, dependencies, and available host primitives
- **Outputs:** Exactly one recommendation, one reason, and one semantic next step
- **Status:** CONTINUE | SUBAGENT | HANDOFF | COMPACT | CLEAR | STOP
- **Consumers:** Any VS workflow at handoff, or a direct "what next?" request
- **Skip conditions:** None at an explicit phase boundary; callers may keep the
  default continue/stop decision internal when the user need not act

## Decide

Read and apply
[`../vs-internal-shared/references/phase-boundaries.md`](../vs-internal-shared/references/phase-boundaries.md).

Use evidence from the current conversation and durable artifacts. Determine:

1. whether the current objective and phase are actually complete
2. whether the next work is authorized and semantically related
3. whether current reasoning is still useful or already replaced by artifacts
4. whether a bounded independent lane can run while the parent continues
5. whether the next work changes owner, workspace, session, or context needs

Do not offer a menu. Choose exactly one recommendation. Do not use a subagent
merely because the parent context is crowded, and do not clear or compact an
unfinished phase whose reasoning has not been preserved.

## Act proportionally

When composed inside a workflow, return the decision to the caller. The caller
may continue or delegate work already authorized by its objective.

When invoked directly, do not begin a new scope merely because it is the
semantic next step. Create a handoff artifact only when the current request
authorizes it; otherwise recommend the exact command, artifact, or fresh-task
action the user should take.

## Output

```text
Recommendation: <continue | use a subagent | hand off | /compact | /clear | stop>
Why: <one evidence-based sentence>
Next: <one semantic workflow, artifact, command, or concrete action>
```

## Output style

Apply the [shared output style](../vs-internal-shared/references/output-style.md)
to every user-facing message.

## Workflow

Direct: emit **Next** only. Composed: return to caller.

**Prev:** phase boundary | workflow handoff
**Next:** done
**Relevant:** none

From vltansky/vs, MIT.

More from vltansky/vs

Signals

Stars
0
Forks
1
Open issues
1
Last commit (this folder)
Aug 21, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from vltansky

Cite this page

vs-next skill. Agent Mart. https://agentmart.dev/skills/vltansky-vs-next (last updated 2026-10-24)