Skip to content
Back to skills

Apex

ASecurity

Run the configurable APEX one-shot workflow. Use when the user invokes $apex or /apex to analyze, plan, implement, optionally review code, and verify a change.

  • 3 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added August 31, 2026
testingcode-reviewgit

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 22, 2026

npx -y skills add twikus/claude-configuration --skill apex --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Apex?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Apex
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/twikus-apex/badge)](https://www.skillsdirectory.com/skills/twikus-apex)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: apex
description: Run the configurable APEX 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>"
---

Run the APEX 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 `$plan` result is the source of truth. Do not combine or reorder stages.

After `$plan` finishes, continue automatically only when `-a` is present. Without `-a`, stop before `$implement`, present the plan and any issue, and wait for approval.

## Flags

- `-a`: autonomous. Continue into implementation without asking. Resolve facts, make safe reversible decisions, and record assumptions in the plan and, when authorized, on the issue. Stop only for missing authority, access, credentials, or an irreversible material decision.
- `-v`: run `$verify` (already the default).
- `-x`: run `$code-review` after implementation and before verification.

Accept grouped forms such as `-axv`.

If `$code-review` or `$verify` changes the implementation, rerun every affected downstream stage. Close a GitHub issue created in this run only after verification passes every acceptance criterion; otherwise leave it open and report the exact blocker.

Preserve `$verify`'s evidence. Show every verification screenshot with Markdown image syntax and its absolute path.

Files in this skill

  • SKILL.md1.9 KB
  • agents/openai.yaml319 B
  • assets/codex-icon.svg364 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…