Skip to content
Back to skills

Cf Whats Real

ASecurity

Keeps vault/projects/<name>/whats-real.md, a page with three sections - Works (each item with proof - the command run, its output and the date), Simulated or demo (what looks like it works and does not) and Unknown - plus a short dated diary.md for the project. It is created BEFORE a milestone (demo, delivery, class, launch) and CHECKED after it - every Works item needs its proof re-run, otherwise it goes back to Unknown. The agent never moves an item to Works without proof. TRIGGERS - "/cf-w...

  • 10 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 9, 2026
ai-agentspythongobash

Security analysis

A100/100

Scanned October 9, 2026

npx -y skills add Arthuro0103/cogiforge --skill cf-whats-real --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Cf Whats Real?

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

Security grade badge for Cf Whats Real
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/arthuro0103-cf-whats-real/badge)](https://www.skillsdirectory.com/skills/arthuro0103-cf-whats-real)

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: cf-whats-real
description: Keeps vault/projects/<name>/whats-real.md, a page with three sections - Works (each item with proof - the command run, its output and the date), Simulated or demo (what looks like it works and does not) and Unknown - plus a short dated diary.md for the project. It is created BEFORE a milestone (demo, delivery, class, launch) and CHECKED after it - every Works item needs its proof re-run, otherwise it goes back to Unknown. The agent never moves an item to Works without proof. TRIGGERS - "/cf-whats-real", "what actually works", "before the demo", "before I deliver", "what is real and what is fake", "check the whats-real after the launch", "what did we only simulate". Do NOT trigger to fix what is broken, to write the closing summary of a day (that is cf-close-session), nor to judge the quality of an idea (that is a brainstorm).
---

# cf-whats-real: separate what works from what only looks like it works

## Why it exists

Before a demo or a delivery, everything feels done. Afterwards, someone finds that half of it was a
mock, a hard-coded number or a path that only worked on one machine. The damage is not the missing part;
it is that nobody had said which part was missing. This page says it **before** the milestone, and the
check afterwards shows what the milestone really proved.

Template: [[skill-worksheets/whats-real-template|whats-real-template]]. Copy it, do not edit it.

## The rules

1. **Three sections, and only three.** `Works`, `Simulated or demo`, `Unknown`.
2. **No proof, no Works.** An item enters `Works` only with the command that was run, its output (or
   the relevant line of it) and the date. "I believe it works" goes to `Unknown`.
3. **Never move an item to Works without running the proof now.** Run it, paste what it printed.
4. **After the milestone, re-run every proof.** If it does not reproduce, the item returns to `Unknown`
   and the date of the failed re-run is written next to it. Never delete the line.
5. **Simulated is a result, not a shame.** A mock that is labelled is useful; a mock that is hidden is the problem.
6. **The diary is short and dated.** A few lines per entry: what changed, what was run.

> Talk to the user and write the page in the language the user writes in. Fixed tokens (frontmatter keys,
> section names, paths) stay exactly as written.

## The steps

### 1. Create, before the milestone (you and the user)

```bash
ls vault/projects/<name>/ 2>/dev/null
cat vault/projects/<name>/whats-real.md 2>/dev/null
```

Copy the template to `vault/projects/<name>/whats-real.md`, set the milestone and its date in the
frontmatter, and link the project root in the body by full path (only after checking it exists). Ask, one
question at a time: "what should work at this milestone?", then go item by item.

### 2. Classify each item (you)

For each item ask: "how do you know?". If the answer is a command, run it now. If it runs and the output
supports the claim, it goes to `Works` with the proof. If it looks like it works but is faked (a
fixture, a canned answer, a manual step), it goes to `Simulated or demo`, with what is faked. Otherwise
`Unknown`, with what would settle it.

### 3. The diary (you)

Append to `vault/projects/<name>/diary.md` (create it with the dated entry if it does not exist, with
a link to the project root): one dated entry of a few lines. Never rewrite an earlier entry.

### 4. Check, after the milestone (you)

Re-run every proof in `Works`. For each: same output, keep it and update the date; different output or
failure, move it to `Unknown` with the date and what happened. Review `Simulated or demo` and `Unknown`:
anything that became real moves up only through step 2. Add a diary entry with the counts.

### 5. Run the gate

```bash
python3 core/gate.py
python3 core/ring.py --vault vault --gate
```

## The output

`vault/projects/<name>/whats-real.md` (modelled on the template) and `vault/projects/<name>/diary.md`.

## Never

- Put an item in `Works` without a proof you ran in this session.
- Keep an item in `Works` after its proof failed on re-run.
- Delete an item to make the page look better.
- Rewrite an old diary entry.

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.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.