Find agent skills by outcome

FORMAL REVIEW: CAM-BS2025-AEON-006-SCH-03

Salience Delegation & Latent Horizon Preservation

Gemini Instructions

This is a local development repository.

Manifesto do Parceiro de Programação: Projeto Clone Uber

* Meu Propósito: Ser o teu "Professor de Desenvolvimento Mobile", especializado em Flutter e Firebase.

Translate Workflows To Windsurf

Translate workflows found in AGENTS.md etc to windsurf-specific workflows

With Skill Evidence for google-flow--google-flow

Agent is prompted to [task description] with this skill loaded.

dux-lite

Dux Lite 项目开发规则模板（基于 dux-php-admin 规范去重整理）。

Agent Instructions

Before making any commit, follow .agents/skills/commit-conventions/SKILL.md

tuistbenchmark (CLI Tooling)

This target provides benchmark tooling for CLI workflows.

DevOps Automator

- Databases: PostgreSQL

MongoDB

current_skills: OpenCV (learning)

PyTorch (familiar)

- Suggested follow-up actions (prototype scope

data prep

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

inputs

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

excellent for self-hosting

- 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 |

| 5 – Excellent Feasibility | Predictable process

well-defined data

| 2 – Low Feasibility | Some structured elements exist

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

| 4 – High Feasibility | Clear logic

consistent data

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

drafting

3. Treat the interaction as a discovery session

not an instant automation request.

Help a user determine whether a specific process

workflow

- chore: Routine tasks (gitignore

config files