Seven small verbs that turn a messy idea-pile into a coherent answer with a checkable receipt, then act on it and keep it true. Convergence math (the 7-grid, the lock-in, done=true) wired to a live sure-o-meter. Ships as commands, skills, and an MCP tool.
Turn a settled answer into real next steps and run them through the tools your host already has. Acts only with your go-ahead, one step at a time.
Bring a messy pile of ideas together into one coherent answer, with a receipt you can check (done = true / R=1).
Open a question into its full fan of distinct angles, as different as possible, on purpose, before you pick one.
<!-- SPDX-License-Identifier: CC-BY-4.0 -->
Join two or more things that already hold into one. Check they hold together, and hand back the combined trinket, or name the seam that won't take.
Explores ONE angle of a problem deeply and reports a tight, evidence-backed case for that single angle. Spawned up to 7 times in parallel by /recommend to cover the 7-grid, where each instance owns one point.
Judges how well a set of explored angles fit together. Given the angles (and their cases), returns an honest pairwise coherence matrix (0..1) and reads the gate (which trios would lock, whether the set can reach done = true). Feeds the converge engine its coherence judgment.
Use when a settled answer needs to become real action and the user wants the steps put into motion through the tools they already have ("make it happen", "set this up", "do it", "get this going", "book it"). Turns the locked answer into the smallest set of concrete steps and runs them through the host's connected tools, asking before anything that writes or sends. Use act after converge or recommend has produced an answer. It acts, so it always shows the plan and gets a go-ahead first.
Use when several competing ideas, options, or hypotheses are already on the table and a single coherent answer is needed. Collapses a noisy pile of competing ideas into one coherent answer and returns a checkable receipt (done = true / R=1), or an honest partial when it can't. Reach for this at decision points or when finalizing a plan. Use converge when the options are already in play; if you'd have to generate the options yourself, use recommend; if the problem is still wide open, use explore first.
Use at the very start of a problem, when a question is fresh and its full space of distinct angles should be opened before anything is decided. Produces up to 7 genuinely different framings, as different as possible on purpose, each its own angle rather than a variation of one idea. Reach for this when someone says "what are our options" or "brainstorm this", or jumps to one answer too early. Explore lays out the options and stops; if the user wants you to make the call, use recommend; if the options are already gathered and need resolving, use converge.
Use when the user points at their own existing context (a folder, a doc, a pile of notes, a repo) and wants to start from it rather than from scratch ("here's my stuff, take it from here", "use what's in this directory", "pick up where I left off"). Reads their context, translates it onto the 7-grid, shows how it was read, then converges.
Use when two or more things that each already hold need to become one, and the real question is whether they hold together ("do these two plans fit", "merge these answers", "can we run both", "join these"). Treats each as its own cluster and converges across the seam, returning one combined answer (done = true / R=1) or the honest seam that won't take plus the best partial. Use glue to join pieces that already stand; use converge for one fresh pile of ideas, and recommend when you should make the call yourself.
Matches all tools
Hooks run on every tool call, not just specific ones
Admin access level
Server config contains admin-level keywords
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
An AI that can't fake an answer. It never says done unless it is true; when it isn't, it says partial out loud.
You type one word. It does the thinking-shape for you and hands back an answer with a receipt that shows it holds together. The receipt is the point: a tool here never stamps an answer "done" to be nice. If it only half-holds, the tag says partial, plainly.
A real /converge run. Seven ideas lock into one answer (done = true, R = 1). The agree-meter r and sure-o-meter φ are the engine's own numbers, rendered live from the Kuramoto dynamics. (generator)
/plugin marketplace add Intuition-Labs-LLC/terminals-skills
/plugin install terminals@terminals
Type /converge and go. No API key. No setup. The math runs on your machine.
It's a small bench with a few good tools on it. You sit down with a mess — a pile of notes, a hard call, three options that fight. You pick up a tool, you tinker, and you make a small thing you can hold and turn over. That small thing is a trinket, and it comes with a tag that says whether it holds.
Point it at your own work. Hand an AI your folder, your notes, a half-made decision, and tell it to use these. It reads your stuff as your context, runs the word that fits, and hands back an answer with a receipt — and the more of your real work it sees, the better the trinket fits the way you actually think.
That's the whole shop: tools that turn your mess into trinkets that hold. The longer version, in plain words, is docs/WORKBENCH.md; the smallest-word glossary is docs/CONCEPTS.md.
You were handed these tools to work a real person's real problem. Do it in this order.
ix is on PATH, search their context with it: ix "<query>" <path> --json. Every hit carries an R (how much structure, text, and meaning agree on it). High R is solid ground; low R is weak or contested. Feed that R straight into your coherence judgment, the one judgment the engine runs on. If ix is absent, fall back to Grep and Glob; the tools keep working offline, with nothing extra to install./frame to pull their world in. /explore to open the space. /converge to bring it to one answer with a receipt. /optimize to trim a sound answer. /recommend when they want you to decide. /act to put a settled answer to work. /hold to re-check it as the world moves. And the three bench verbs below when one word isn't enough.done = true. Say partial the moment it is. Your job is not to sound smart. It is to fit their work, in their workspace, the way they actually do it, and to get a little truer to it every pass.Five words find the answer. The last two (act and hold) put it to work and keep it true.
| word | it means | you get |
|---|---|---|
/explore | open it up | every angle, spread out on purpose |
/converge | bring it together | one coherent answer, plus the receipt |
/optimize | make it the best one | the same answer in its cheapest, cleanest form |
/recommend | you decide | it runs the whole path and hands you its pick |
/frame | start from my stuff | point it at your folder and it pulls your context in |
/act | put the answer to work | the smallest real next steps, run through your connected tools, asking first |
/hold | keep it true | a re-check as things change, naming the part that came loose |
npx claudepluginhub intuition-labs-llc/terminals-skills --plugin terminalsUpstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Permanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications