Skip to content
Agentmart
Submit a tool

reach-bug

Use when a check or qualification fails, a grade comes back with a failed case, or the student says something gives the wrong answer.

Skill by SteveBenner

IndexedMIT
Write a review

Chalk Labs

Need help deploying reach-bug?

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-bug

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 a check or qualification fails, a grade comes back with a failed case, or the student says something gives the wrong answer. The bug flow for one slice, without git.

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-bug

What's inside

Skill name
reach-bug
Description
Use when a check or qualification fails, a grade comes back with a failed case, or the student says something gives the wrong answer. The bug flow for one slice, without git.
License
MIT
Instructions
11 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

The bug flow for one slice: the smallest fix, proven by the same run. Work the steps in order. Never run git or start a repository; overwrite files in place and use `reach checkpoint` as the history.

1. **Name it exactly.** The QF code and scenario name from `reach qualify`, the finding id and file from `reach check --format agent`, the failed case on a grade receipt, or the student's own description of the wrong answer. Quote it word for word in your notes; tell the student about it only in business terms.
2. **Reproduce it** with the same command before changing anything. A wrong answer the student reports: write a scenario under qualify/features that shows it and watch it fail. For `QF-UNCOVERED`, read the returned `steps.coverage.rows` and `next`: behavior checks have not run yet, and the assistant owns the missing technical scenarios. The student's explanations and contributions remain their own work.
3. `reach checkpoint save --note "before fix: <name>"`.
4. **Show it in a scenario.** Add or correct the scenario that captures the case, with the slice's tag, so the fix is proven and stays proven. For `QF-UNCOVERED`, write a scenario for each missing name, with exactly that name and tag, from the agreed plan and course contract. Ask the student only for a business expectation that is still missing. Follow the ladder's current limits and stop conditions; continue to the requalification step instead of finishing with a readiness summary.
5. **The smallest fix** in the owned files, reading each file before you overwrite it. Do not rewrite a file or fix things that were not named.
6. `reach check --format agent`, then `reach qualify`.
7. **Follow the ladder** `reach qualify` prints (the TOZERO directive). When findings grow instead of shrinking, `reach checkpoint restore <n>` to the last good one and report.
8. **Passed:** `reach checkpoint save --note "fixed: <name>"` and tell the student in at most four plain lines what now works, with their case.

From SteveBenner/rEach, MIT.

More from SteveBenner/rEach

Signals

Stars
0
Forks
0
Open issues
2
Last commit (this folder)
Oct 11, 2026 (19 days ago)
Last release
Oct 22, 2026 (9 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-bug 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-bug?

CallGet help
Cite this page

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