Menú
Analyze and optimize a prompt for the DQIII8 ecosystem. Classifies prompt type (LLM routing, image gen, TTS, agent instruction, Telegram), scores 5 dimensions, produces optimized version.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
| name | prompt-optimize |
| description | Analyze and optimize a prompt for the DQIII8 ecosystem. Classifies prompt type (LLM routing, image gen, TTS, agent instruction, Telegram), scores 5 dimensions, produces optimized version. |
| command | /prompt-optimize |
| allowed-tools | ["Read"] |
| user-invocable | true |
Analyze and optimize a prompt for maximum effectiveness in the DQIII8 ecosystem (LLM routing, scene director, TTS, Telegram bot, or agent instructions).
/prompt-optimize [the prompt text or file path]
Given the prompt in $ARGUMENTS:
| Dimension | What to check |
|---|---|
| Clarity | Unambiguous intent, no vague instructions |
| Specificity | Concrete examples, exact formats, numbers |
| Conciseness | No filler words, no redundant instructions |
| Output contract | Explicit format (JSON/markdown/text) stated |
| Edge cases | Handles empty input, failure, ambiguity |
Apply these DQIII8-specific improvements:
For LLM prompts:
Output ONLY JSON (no markdown): if structured output neededFor image prompts (fal.ai):
no text no watermarks no logosFor TTS narration:
For agent instructions:
## When NOT to use section if missingFor Telegram messages:
*bold* for key metrics, `code` for commandsPROMPT OPTIMIZATION REPORT
===========================
Type: [classified type]
Original score: [X]/50
Scores:
Clarity: [X]/10 — [one-line finding]
Specificity: [X]/10 — [one-line finding]
Conciseness: [X]/10 — [one-line finding]
Output contract:[X]/10 — [one-line finding]
Edge cases: [X]/10 — [one-line finding]
Key issues:
1. [issue + fix]
2. [issue + fix]
Optimized prompt:
---
[full optimized prompt text]
---
Optimized score: [X]/50
$ARGUMENTS
Skill para despachar tareas a agentes NIM/Groq/GitHub desde esta sesión CC. Habilita el Hermes Work Loop — CC orquesta, agentes B/B+ ejecutan.
Upload DOCX files from intl-reports to Google Drive Carpeta 3 using rclone. Zero reasoning — action only.
Defensive security — reads red-team findings and systematically patches every vulnerability. Hardens code, adds input validation, fixes permissions, and verifies each fix. Works from the latest red-team report.
Launch an efficiency audit on modified Python files using Aider + Gemini 2.0 Flash. Analyzes bugs, efficiency, readability, and security. Saves report to database/audit_reports/.
Genera y entrega informes de internacionalización (Diagnóstico + Plan) para pymes españolas. Batch pipeline via Orchestrator v4 (core.cli) desde tmux externo. NUNCA empresa-a-empresa con Agent Haiku manual.
Adversarial security testing — attempts to break the codebase using real hacker techniques. Tests OWASP Top 10, prompt injection, MCP poisoning, dependency attacks, auth bypass, and vibe-coding-specific patterns. Generates exploit report. NEVER auto-invoked — user must explicitly request.