How do I install the Dl Capitalize skill?

Run npx skillmds@latest add thewatcher01/dl-capitalize in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Dl Capitalize safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Dl Capitalize?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Dl Capitalize free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Dl Capitalize?

TheWatcher01 (@thewatcher01) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Dl Capitalize by thewatcher01 — Agent Skill

Name: Dl Capitalize
Rating: 4.2 (5 reviews)
Author: TheWatcher01

File contents

Capitalisation Workflow → .workflow.md

Tu vas cristalliser un workflow repetitif en fichier .workflow.md pour le framework chainskills.

Etapes

Analyser la session : Identifier les actions repetees (outils utilises, fichiers modifies, patterns)
Extraire les invariants : Ce qui est identique a chaque execution (structure, validations, patterns)
Extraire les variables : Ce qui change (noms, chemins, configs) → deviennent des inputs:
Cristalliser : Ecrire le .workflow.md avec le format ci-dessous

Format .workflow.md obligatoire

---
name: <kebab-case>
description: <ce que fait le workflow>
version: 0.1.0
inputs:
  - name: <param>
    type: string|number|boolean|list
    required: true|false
    description: <description>
outputs:
  - name: <result>
    type: string|object
    description: <description>
tags: [datalake, <category>]
metadata:
  author: TheWatcher01
  project: datalake-souverain
---

# Step N — Titre

Description en langage naturel.

@call shell.exec("commande $variable") -> $resultat
@agent copilot: "Prompt avec $context" -> $output
@assert $condition "message erreur"
@parallel:
## Branche A
@call ...
## Branche B
@call ...

Directives disponibles

Directive	Usage
`@call shell.exec("cmd") -> $var`	Commande shell
`@agent copilot: "prompt" -> $var`	Deleguer au LLM
`@if $var == "val":`	Condition
`@for $item in $list:`	Boucle
`@parallel:`	Execution parallele (DAG)
`@try: ... @on-error: log and continue`	Gestion erreurs
`@assert $var != "" "msg"`	Validation
`@output $var -> $result`	Declarer un output
`@workflow sub-workflow:`	Composer des sous-workflows

Regles de capitalisation

Copier+adapter > generer : Toujours referencer les templates existants dans le workflow
Deterministe > creatif : Les steps shell sont deterministes, @agent seulement pour les taches creatives
Paralleliser : Utiliser @parallel: pour les steps independants (ex: lire template + lire config)
Provenance : Inclure lineage_run_id dans tout workflow data
Testable : Chaque workflow doit pouvoir etre --dry-run

Ou sauvegarder

Ecrire dans : ~/projects/chainskills/cli-mcp-core/templates/datalake/<name>.workflow.md

Workflows existants (ne pas dupliquer)

add-crawler — Nouveau crawler (AsyncGenerator + TokenBucket + CircuitBreaker)
add-api-route — Nouvelle route Hono
add-data-import — Import dataset (CSV/JSON/API)
fix-type-errors — Correction erreurs TypeScript
cross-ref-source — Cross-reference entre sources
verify-data — Verification ISO 8000-8 (dans templates/data/)

---name: dl-capitalizedescription: Capitalise un workflow repetitif en fichier .workflow.md chainskills. Detecte les patterns, cristallise les etapes, rend executable par Haiku.---# Capitalisation Workflow → .workflow.mdTu vas cristalliser un workflow repetitif en fichier `.workflow.md` pour le framework chainskills.## Etapes1. **Analyser la session** : Identifier les actions repetees (outils utilises, fichiers modifies, patterns)2. **Extraire les invariants** : Ce qui est identique a chaque execution (structure, validations, patterns)3. **Extraire les variables** : Ce qui change (noms, chemins, configs) → deviennent des `inputs:`4. **Cristalliser** : Ecrire le .workflow.md avec le format ci-dessous## Format .workflow.md obligatoire```markdown---name: <kebab-case>description: <ce que fait le workflow>version: 0.1.0inputs: - name: <param> type: string|number|boolean|list required: true|false description: <description>outputs: - name: <result> type: string|object description: <description>tags: [datalake, <category>]metadata: author: TheWatcher01 project: datalake-souverain---# Step N — TitreDescription en langage naturel.@call shell.exec("commande $variable") -> $resultat@agent copilot: "Prompt avec $context" -> $output@assert $condition "message erreur"@parallel:## Branche A@call ...## Branche B@call ...```## Directives disponibles| Directive | Usage ||-----------|-------|| `@call shell.exec("cmd") -> $var` | Commande shell || `@agent copilot: "prompt" -> $var` | Deleguer au LLM || `@if $var == "val":` | Condition || `@for $item in $list:` | Boucle || `@parallel:` | Execution parallele (DAG) || `@try: ... @on-error: log and continue` | Gestion erreurs || `@assert $var != "" "msg"` | Validation || `@output $var -> $result` | Declarer un output || `@workflow sub-workflow:` | Composer des sous-workflows |## Regles de capitalisation1. **Copier+adapter > generer** : Toujours referencer les templates existants dans le workflow2. **Deterministe > creatif** : Les steps shell sont deterministes, `@agent` seulement pour les taches creatives3. **Paralleliser** : Utiliser `@parallel:` pour les steps independants (ex: lire template + lire config)4. **Provenance** : Inclure `lineage_run_id` dans tout workflow data5. **Testable** : Chaque workflow doit pouvoir etre `--dry-run`## Ou sauvegarderEcrire dans : `~/projects/chainskills/cli-mcp-core/templates/datalake/<name>.workflow.md`## Workflows existants (ne pas dupliquer)- `add-crawler` — Nouveau crawler (AsyncGenerator + TokenBucket + CircuitBreaker)- `add-api-route` — Nouvelle route Hono- `add-data-import` — Import dataset (CSV/JSON/API)- `fix-type-errors` — Correction erreurs TypeScript- `cross-ref-source` — Cross-reference entre sources- `verify-data` — Verification ISO 8000-8 (dans templates/data/)

Frequently asked questions

Dl Capitalize