agentleFS
Sign inStart free
fusebase-devfusebase-dev/fusebase-flow/.cursor/rules/fusebase-flow-validation.mdc

fusebase-flow / rules

Cursor rule9 starsChanged 27 days agoMITView on GitHub

Fusebase Flow validation/verification phase rules. Use when running gate checks, reviewing diffs, or verifying smoke prompts.

The file

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    If any field is missing, redirect: "Gate report missing <field>.

    line 13
  2. Distinctive:

    When operator describes a single observed failure ("the system did X"):

    line 19
  3. Distinctive:

    Don't draft a fix immediately.

    line 21
  4. Very specific:

    When verification-gate.md defines numbered S1..Sn:

    line 30
  5. Distinctive:

    Verify the operator-visible outcome, not only supporting checks

    line 33
  6. Very specific:

    If below threshold or end-to-end smoke is not feasible, do NOT mark spec DONE.

    line 37

≈ “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-validation.mdc
---
description: Fusebase Flow validation/verification phase rules. Use when running gate checks, reviewing diffs, or verifying smoke prompts.
globs: ["docs/specs/*/verification-gate.md", "docs/tmp/handoff/*-smoke/**", "**/test/**", "**/__tests__/**"]
alwaysApply: false
---

# Fusebase Flow — validation rules

## Gate report (AI Developer-side)

Required fields are canonical in `policies/gate-contracts.yml: gate_report` (machine schema); the producer template is `templates/gate-report.md`. Do not restate the field list.

If any field is missing, redirect: "Gate report missing <field>. Per FR-05, complete reports only. Re-run."

Rows are AI Developer PASS/FAIL claims; only the Product Owner side issues the verdict, after its own recount, and gate rounds are bounded (default 2, then `BLOCKED-AT-gate-budget`) per `workflows/verification-gate.md`.

## Reproducibility before fix (FR-10)

When operator describes a single observed failure ("the system did X"):

1. Don't draft a fix immediately.
2. Reproduce 3 times under the same conditions.
3. Outcomes:
   - 3/3 reproduce → systemic; draft fix
   - 1/3 or 2/3 → deterministic app code = real race/state bug unless Explained-external; LLM/agent behavior = model-variance no-op close (`flow-skills/validation-and-qa/SKILL.md` Sub-mode C)
   - 0/3 → close as no-op-needed

## Smoke prompts (post-deploy)

When `verification-gate.md` defines numbered S1..Sn:

- Follow `flow-skills/smoke-testing/SKILL.md`
- Verify the operator-visible outcome, not only supporting checks
- Inspect the ground-truth diagnostic surface named by S<n>
- Persist evidence to `docs/tmp/handoff/<date>-<slug>-smoke/`
- Compute pass ratio against gate contract threshold
- If below threshold or end-to-end smoke is not feasible, do NOT mark spec DONE; surface failure or `PENDING-OPERATOR-SMOKE` with concrete `S<n> observed Y, expected Z` / missing prerequisite

## What this rule does NOT do

- Approve deploy (that's `release-deploy-reporting`)
- Auto-fix lint/typecheck errors
- Skip reproducibility-before-fix

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.

Check your setup, free