Source Command Audit Prose logo

Source Command Audit Prose

CommunityPopular
FlorianBruniaux
source-command-audit-prose

Add descriptive prose to bare sections in whitepapers (FR + EN) using 9 parallel agents

Overview

PublisherFlorianBruniaux
Repositoryclaude-code-ultimate-guide
Skill namesource-command-audit-prose
Stars
6.1K
Forks
807
Bundled files
Instructions only
LicenseCC-BY-SA-4.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by FlorianBruniaux on GitHub. Read the source before you install it.

Installation

Install the Source Command Audit Prose AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/FlorianBruniaux/claude-code-ultimate-guide/tree/main/.agents/skills/source-command-audit-prose
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/FlorianBruniaux/claude-code-ultimate-guide.git /tmp/claude-code-ultimate-guide
mkdir -p .claude/skills
cp -r /tmp/claude-code-ultimate-guide/.agents/skills/source-command-audit-prose .claude/skills/source-command-audit-prose
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Source Command Audit Prose in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Source Command Audit Prose on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Source Command Audit Prose is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

source-command-audit-prose

Use this skill when the user asks to run the migrated source command audit-prose.

Command Template

Audit Prose — Enrichissement des Sections Bare

Ajoute de la prose descriptive (2-4 phrases) devant chaque section "bare" des whitepapers. Une section "bare" = heading suivi directement d'une table, code block, ou liste sans paragraphe d'intro.

Usage

/audit-prose        # Tous les WPs (00-08), FR + EN
/audit-prose 07     # WP07 seulement (FR + EN)
/audit-prose 03 07  # WP03 et WP07 seulement

Définition "Bare Section"

Section (## ou ###) suivie DIRECTEMENT par :

  • Une table Markdown (|)
  • Un code block (```)
  • Une liste à puces (-) ou numérotée

Sans paragraphe d'introduction (minimum 2 phrases).

Règles Strictes (non négociables)

  1. NE JAMAIS terminer un paragraphe par : avant une liste à puces → bug Typst qui fusionne les bullets en inline
  2. NE PAS inventer de chiffres, métriques, ou citations
  3. NE PAS ajouter de callouts (:::), Mermaid, ou scénarios persona
  4. NE PAS dépasser +15% de lignes par fichier
  5. Versions maintenues à 3.27.6
  6. Prose EN = anglais naturel, pas de traduction mot-à-mot du FR
  7. Chaque ajout = minimum 2 phrases, maximum 4 phrases

Exemple de Transformation

AVANT :
### Exit Codes
| Code | Signification |
|------|---------------|
| 0    | Autorisé      |

APRÈS :
### Exit Codes
Les hooks communiquent avec Codex via des codes de sortie standardisés.
C'est le mécanisme qui permet de bloquer ou autoriser une action en temps réel,
sans intervention humaine.

| Code | Signification |
|------|---------------|
| 0    | Autorisé      |

Workflow : 9 Agents Parallèles

Sources de contexte

  • machine-readable/reference.yaml : mapping sujets → lignes du guide principal
  • guide/ultimate-guide.md : guide de référence (aux lignes indiquées dans reference.yaml)
  • Connaissance Codex (v3.27.6, MCP, hooks, agent teams, etc.)

Mapping WPs

#FREN~Bare sections
00whitepapers/fr/00-introduction-serie.qmdwhitepapers/en/00-series-introduction.qmd~10
01whitepapers/fr/01-prompts-efficaces.qmdwhitepapers/en/01-effective-prompts.qmd~14
02whitepapers/fr/02-personnalisation.qmdwhitepapers/en/02-customization.qmd~18
03whitepapers/fr/03-securite.qmdwhitepapers/en/03-security.qmd~20
04whitepapers/fr/04-architecture.qmdwhitepapers/en/04-architecture.qmd~14
05whitepapers/fr/05-equipe.qmdwhitepapers/en/05-team.qmd~20
06whitepapers/fr/06-privacy.qmdwhitepapers/en/06-privacy.qmd~12
07whitepapers/fr/07-guide-reference.qmdwhitepapers/en/07-reference-guide.qmd~36
08whitepapers/fr/08-agent-teams.qmdwhitepapers/en/08-agent-teams.qmd~16

Exécution

Si argument = numéro WP spécifique : lancer 1 agent pour cette paire FR/EN uniquement.

Si pas d'argument (ou "all") : créer une équipe audit-prose-{date} et lancer 9 agents en parallèle.

Pour chaque WP ciblé, lancer un agent avec ce prompt :

Tu es un éditeur technique senior. Ta mission : ajouter de la PROSE DESCRIPTIVE
aux sections "bare" du whitepaper [XX - Titre].

## Définition "bare section"
Section (## ou ###) suivie DIRECTEMENT par une table, un code block, ou une
liste à puces, SANS paragraphe d'introduction (2+ phrases).

## Ce que tu fais
Pour CHAQUE bare section identifiée :
1. Ajouter 2-4 phrases AVANT le contenu structuré
2. Chaque phrase apporte une INFO NOUVELLE : le pourquoi, le quand utiliser,
   les implications, le choix de design, ou le contexte pratique
3. PAS de reformulation de ce qui suit — COMPLETER avec du contexte manquant

## Sources de contexte
- `machine-readable/reference.yaml` : mapping complet sujets → guide principal
- Ta connaissance de Codex (v3.27.6, MCP, hooks, agent teams, etc.)

## Règles strictes
- NE PAS inventer de chiffres, métriques, ou citations
- NE PAS ajouter de callouts (:::), Mermaid, ou scénarios persona
- NE JAMAIS terminer un paragraphe par ":" avant une liste à puces (bug Typst)
- NE PAS dépasser +15% de lignes
- Garder les versions à 3.27.6
- Prose EN = anglais naturel, pas traduction mot-à-mot
- Chaque ajout = minimum 2 phrases, maximum 4 phrases

## Processus
1. LIRE reference.yaml (sections pertinentes au WP)
2. LIRE le WP FR complet
3. IDENTIFIER toutes les bare sections
4. ÉDITER le fichier FR (Edit tool, section par section)
5. LIRE le WP EN complet
6. ÉDITER le fichier EN
7. VÉRIFIER : aucun paragraphe ne finit par ":" avant une liste
8. Rapporter : sections traitées, delta lignes FR/EN

Attente et Reporting

Attendre les retours de tous les agents. Pour chaque WP completed, afficher :

WP[XX] ✅ — [N] sections, +[X]% FR/EN

Shutdown

Quand tous les agents ont terminé, envoyer shutdown_request à chacun, puis TeamDelete.

Vérification Post-Enrichissement

bash
# 1. Delta lignes (max +15%)
for f in whitepapers/fr/*.qmd; do
  echo "$(wc -l < "$f") $(basename $f)"
done

# 2. Vérifier bug Typst : paragraphe finissant par ":" suivi de liste
grep -n ':$' whitepapers/fr/*.qmd whitepapers/en/*.qmd | head -20

# 3. Versions intactes
grep "^version:" whitepapers/fr/*.qmd whitepapers/en/*.qmd

# 4. Rebuild PDFs
for f in whitepapers/fr/*.qmd whitepapers/en/*.qmd; do
  quarto render "$f"
done

Budget Lignes par WP

WPBudget max FRBudget max EN
00+54 lignes+54 lignes
01+142 lignes+142 lignes
02+189 lignes+189 lignes
03+145 lignes+145 lignes
04+120 lignes+120 lignes
05+143 lignes+143 lignes
06+79 lignes+79 lignes
07+281 lignes+281 lignes
08+241 lignes+241 lignes

Frequently asked questions

What does the Source Command Audit Prose AI skill do?

Add descriptive prose to bare sections in whitepapers (FR + EN) using 9 parallel agents

Why use Source Command Audit Prose on TypingMind?

Because you install it once and use it with any model. Source Command Audit Prose is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Source Command Audit Prose in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/FlorianBruniaux/claude-code-ultimate-guide/tree/main/.agents/skills/source-command-audit-prose. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Source Command Audit Prose?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Source Command Audit Prose?

As many as you like. As long as a model supports skills, you can use Source Command Audit Prose with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Source Command Audit Prose AI skill free?

Yes. It is published on GitHub by FlorianBruniaux under the CC-BY-SA-4.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇