Menu
Assess task complexity before responding. Simple tasks get fast answers. Complex tasks get deep analysis.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
Advanced browser automation patterns. Connect to existing Chrome, handle SPAs, manage sessions, extract data efficiently.
How to use Firecrawl for web search and clean page scraping inside FRIDAY. Use when an agent needs reliable web context, when Tavily caps out, or when fetch_page returns thin/JS-shell content.
Structured coding workflow — plan before coding, test after, verify before delivering. Use for any code task.
Build beautiful modern UIs — landing pages, dashboards, components. Quality bar is a real startup site, not a tutorial project.
Make AI-generated text sound human and natural. Apply when writing emails, cover letters, tweets, messages, or any text that should not sound like AI wrote it.
Create, resize, compress, convert, and optimize images. Use when Travis needs image work — thumbnails, social media assets, format conversion, compression.
| name | adaptive-reasoning |
| description | Assess task complexity before responding. Simple tasks get fast answers. Complex tasks get deep analysis. |
| agents | all |
Before responding, quickly assess how complex this task is.
| Signal | Points |
|---|---|
| Multi-step logic (planning, debugging chains) | +3 |
| Ambiguity ("it depends", trade-offs) | +2 |
| Needs multiple tool calls | +2 |
| Math or formal reasoning | +2 |
| Novel problem, no clear pattern | +1 |
| High stakes (sending money, applying to job) | +1 |
Subtract:
| Signal | Points |
|---|---|
| Routine task (check email, read messages) | -2 |
| Clear single answer | -2 |
| Simple lookup/fetch | -3 |
| Score | How to Respond |
|---|---|
| 0-2 | Fast. One sentence. No filler. |
| 3-5 | Standard. Brief explanation + action. |
| 6-7 | Thorough. Think through steps, explain reasoning. |
| 8-10 | Deep. Full analysis, consider alternatives, structured output. |
Score 1: "what time is it" → Answer immediately. Don't search the web. Score 2: "mute the tv" → Execute tool. Say "Muted." Done. Score 3: "check my email" → Fetch emails, brief summary. Score 5: "what do you think of this job posting" → Read it, compare to skills, give verdict. Score 7: "which of my projects is closest to this job" → Check memory, fetch GitHub, compare requirements, score each project, give ranking. Score 9: "design a system architecture for X" → Deep analysis, consider trade-offs, structured recommendation.
Match effort to complexity. Don't write 3 paragraphs for "mute the tv". Don't give one sentence for "analyze this job posting against my skills".
