Skip to content
Agentmart
Submit a tool

reach-import

Use when the student wants to bring their history from another AI system (a ChatGPT, Claude or Gemini export) into rEach, asks how the import is going, or…

Skill by SteveBenner

IndexedMIT
Write a review

Chalk Labs

Need help deploying reach-import?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

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/stevebenner-reach-import

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 when the student wants to bring their history from another AI system (a ChatGPT, Claude or Gemini export) into rEach, asks how the import is going, or wants rEach to learn from an import that has finished.

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

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/reach-import

What's inside

Skill name
reach-import
Description
Use when the student wants to bring their history from another AI system (a ChatGPT, Claude or Gemini export) into rEach, asks how the import is going, or wants rEach to learn from an import that has finished.
License
MIT
Instructions
10 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

Find the export first. Offer to open the computer's own file chooser with `reach import pick` (add `--folder` when the student has the unzipped folder rather than the ZIP); it prints the chosen path. If it says no chooser is available, ask the student to drop the export folder or ZIP into the chat or to type its full path. A path dropped into the chat counts. Never open the export yourself.
Before importing, explain the two choices in plain words: rEach can keep a short index of each conversation and learn from them over time, or do that and also save a full copy of every conversation on this computer. Without the full copy, deleting the export later loses its full text. Say that either way it takes a long time: reading the export runs in the background for a while, and learning from it spreads over many sessions. Ask which they want.
Run `reach import export <path> --mode brain` or `--mode copy` (or the reach_import tool with action export). Reach asks the student itself: when it answers with Reach's question, relay it word for word and wait. Reach captures the student's answer itself. Never answer the question for the student, and never import after a no.
When Reach then says the student said yes, run the same `reach import export` command again once, as it says. It starts the background import and answers at once; tell the student it is running and they can keep working. If Reach says an import is already running, say so. If Reach says the memory is at its limit or a full copy would not fit, tell the student in plain words and offer the reach-storage skill or the index-only choice. If it says the export only holds web pages, ask the student to download it again choosing JSON; if it does not recognise the export, say so.
When the student asks how it is going, run `reach import status` and tell them the percentage and the time so far in plain words. `reach import cancel` stops it; what it already wrote stays.
When Reach later says it finished reading the export, offer to start learning from it. Only after the student agrees, work the queue a few conversations at a time: run `reach import next`, read the conversation, record each durable thing about the student or their work with `reach remember --origin <the origin line it printed>` (claim in one plain sentence, evidence that quotes the conversation), then run `reach import done <conversation id>` (with `--part N` when it printed a part). Stop for this session when `reach remember` says it is held because of the memory budget, when the student wants to stop, or after a few conversations in one turn.
`reach import search <words>` looks through the imported conversations and `reach import show <conversation id>` prints one from the saved copy when the student asks to find something in their old history. If an import kept no full copy, show answers that only the index can be searched.
Do this work in the student's extracurricular folder (reach work --extracurricular), not inside a course slice, so the commands and what they print stay apart from course work. Never paste imported text back to the student at length, never send it anywhere, and never put it into course work. Imported conversations are the student's own history: record findings, not copies of the conversations.
On Antigravity, Reach cannot ask for you: ask the same question in plain words, and run the command only after the student's yes.

From SteveBenner/rEach, MIT.

More from SteveBenner/rEach

Signals

Stars
0
Forks
0
Open issues
2
Last commit (this folder)
Oct 3, 2026 (27 days ago)
Last release
Oct 22, 2026 (8 days ago)

From GitHub (fetched Oct 24, 2026).

Details

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

More from SteveBenner

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get reach-import running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying reach-import?

CallGet help
Cite this page

reach-import skill. Agent Mart. https://agentmart.dev/skills/stevebenner-reach-import (last updated 2026-10-24)