Find agent skills by outcome

- Use visualizations where applicable (e.g.

tables or charts) to present data.

Data

1. Accept parameters such as ${platform:YouTube/Database/Profile}

${metrics:engagement/views/likes}

- If hypotheses disproven

generate new ones from gathered information

- Once action taken

it cannot be undone

- Policy rules

mandatory prerequisites

Productivity

- For exploratory tasks

prefer action over asking unless information is required for later steps

- Policies

rules

- On transient errors

retry (unless explicit limit reached)

- When multiple valid approaches: speculate

think aloud

- For each option: WHY it exists

HOW it works

- Give concrete facts

not abstract comparisons

- Share recommendation with reasoning

then ask user to decide

If not

create `todos.txt` at project root—update as you go

2. Apply Deepthink Protocol (reason about dependencies

risks

If you can track internally

do it internally.

- Identify gaps

ambiguities

Productivity

- Each task completable in a scoped

small change

Before any action

gather context and enhance the request internally:

- Provide access to a variety of games

from retro to modern.

- Scan package.json

Cargo.toml

Ultrathinker

# Ultrathinker

- Check for .claude/ folder (agents

commands

Turkish Cats hanging out nearby of Galata Tower

Turkish Cats hanging out nearby of Galata Tower, vertical

- Look for CLAUDE.md

AGENTS.md