agentleFS
fusebase-devfusebase-dev/fusebase-flow/.cursor/rules/fusebase-flow-implementation.mdc

fusebase-flow / rules

Cursor rule9 starsChanged 27 days agoMITView on GitHub

Fusebase Flow implementation phase rules. Use when editing source code per a locked spec.

  • Deletes or force-pushes
  • Commits and pushes

The file

We split this fusebase-flow-implementation.mdc into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use only when in AI Developer role with a locked spec.

    line 9
  2. Distinctive:

    If any check fails, STOP and ask the operator.

    line 19
  3. Very specific:

    Commit message format: <type>(<scope>): T<n> <one-liner>

    line 26
  4. Very specific:

    When tasks T<first>..T<gate> are committed, produce the gate report per docs/specs/<slug>/verification-gate.md.

    line 48

≈ “Common” is an estimate: at least 1 in 5 of the 6,668 Cursor rule files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
fusebase-flow-implementation.mdc
---
description: Fusebase Flow implementation phase rules. Use when editing source code per a locked spec.
globs: ["**/*.{ts,tsx,js,jsx,py,go,rs,java,rb,php,sql}"]
alwaysApply: false
---

# Fusebase Flow — AI Developer discipline

Use only when in AI Developer role with a locked spec.

## Self-check before any code edit

- [ ] Spec exists at `docs/specs/<slug>/spec.md` and is referenced
- [ ] Decisions are LOCKED in `docs/specs/<slug>/decisions.md`
- [ ] Tasks list exists with the current T-number declared
- [ ] Pre-task git checkpoint clean (`git status --short` empty)
- [ ] Worker-undisturbed paths from `policies/protected-paths.yml` will show empty diff for this task

If any check fails, STOP and ask the operator.

## Per-task discipline

- One task = one commit (FR-03)
- Lint + typecheck clean per code commit; declared docs-only commits skip them (FR-13)
- Stage files explicitly by name — never `git add -A` / `git add .` (FR-06)
- Commit message format: `<type>(<scope>): T<n> <one-liner>`
  - `feat(spa): T17 ...`
  - `fix(extension): T18 ...`
  - `test(backend): T19 ...`
  - `docs(post-deploy): T20 ...`

## Pre-commit attestation (output before each commit)

```
T<n> pre-commit check:
☐ Lint clean
☐ Typecheck clean
☐ Worker-undisturbed unchanged
☐ One task scope (no bundling)
☐ No TODO/FIXME/WIP markers
☐ Commit message cites T<n>

→ Committing T<n>: <scope>
```

## Stop at gate

When tasks T<first>..T<gate> are committed, produce the gate report per `docs/specs/<slug>/verification-gate.md`. Do NOT proceed to T<deploy>. Wait for an explicit deploy handoff (FR-05).

## Forbidden without operator confirmation

`rm -rf`, `git push --force`, `git reset --hard`, `git checkout -- .`, `git clean -fdx`, `git add .`, `git add -A`, `--no-verify`. Full list at `policies/command-policy.yml`.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.