Apex

---
name: apex
description: Run the configurable Melvyn one-shot workflow. Use when the user invokes $apex or /apex to analyze, plan, implement, optionally review code, and verify a change.
disable-model-invocation: true
argument-hint: "[-a] [-v] [-x] <request>"
---

# Apex

Run the Melvyn workflow by calling its skills in this order:

1. `$analyze`
2. `$plan`
3. `$implement`
4. `$code-review` only with `-x`
5. `$verify` by default; `-v` explicitly requests the default

Pass each result to the next skill. The implementation plan returned by `$plan` is the source of truth. `$plan` creates a durable GitHub issue when requested, required by repository rules, or when its size and coordination rules classify the plan as large. Do not combine or reorder stages.

## Flags

- `-a`: autonomous mode. Do not pause after analysis or planning for validation. Resolve facts, make safe reversible decisions, and record assumptions in the current plan and, when an issue is in use and mutation is authorized, on that issue. Stop only for missing authority, access, credentials, or an irreversible material decision.
- `-v`: run `$verify`. Verification is already on by default.
- `-x`: run `$code-review` after implementation and before verification.

Accept grouped forms such as `-axv`. No other workflow flags are supported.

If `$code-review` or `$verify` changes the implementation, rerun every affected downstream stage. When this run created a GitHub issue, close it only after verification passes every acceptance criterion; otherwise leave it open and report the exact blocker. Without an issue, complete the workflow directly from the plan and verification result.

Preserve `$verify`'s evidence in the final response. When verification captures screenshots, show every screenshot directly with Markdown image syntax and its absolute path; never replace the images with paths, links, a report, or a summary.