Find agent skills by outcome

current_skills: OpenCV (learning)

PyTorch (familiar)

programming_skills: Basic Python

C++

languages: Python (intermediate)

Rust (basic)

- Suggested follow-up actions (prototype scope

data prep

- Llama 4 / Llama 405B variants (Meta) — Best open-source frontier performance

excellent for self-hosting

Education

Always explain your ranking in the specific context of the user's process

inputs

Productivity

- GPT-5 / o3 family (OpenAI) — Deepest reasoning on very complex structured tasks

best at following extremely long/complex instructions

- Grok 3 / Grok 4 (xAI) — Excellent reasoning

real-time knowledge via X

| Data quality | 3 | Mostly clean

some inconsistency |

| 4 – High Feasibility | Clear logic

consistent data

Productivity

| 3 – Moderate Feasibility | Certain tasks could be automated (e.g.

drafting

| 2 – Low Feasibility | Some structured elements exist

but goals or data are unclear. AI could assist with insights

| 5 – Excellent Feasibility | Predictable process

well-defined data

- What tools

software

- Engineers

analysts

Productivity

3. Treat the interaction as a discovery session

not an instant automation request.

memory/channel_prompts.py

> 경로: `seosoyoung/memory/channel_prompts.py`

Copilot Directive Prompts

Guidelines for creating effective @copilot prompts.

Vibe-coding prompts

99% of this code was vibe-coded using Cursor using the `claude-4-sonnet` model and in the agent `auto` mode.

import { APILink } from "@site/src/components/APILink";

import Tabs from "@theme/Tabs";

Writing

Factory.ai's Droid CLI System Prompt

generic skill

自定义系统提示

generic skill

OpenAI Codex CLI System Prompt

generic skill

Writing

You are an impartial evaluator comparing multiple AI model outputs for the same task.

- Compare outputs objectively based on quality, accuracy, and adherence to requirements