Find agent skills by outcome

- Never silently assume requirements

APIs

If the answer is anything other than append-only

the agent must STOP and ask for confirmation.

Before modifying any md file

the agent must internally verify:

- Am I rewriting for clarity

cleanup

- Phrases like “go ahead”

“implement”

- Until approval is given

limit output to analysis

Coding & Debugging

- This includes example snippets

pseudo-code

(e.g. Antigravity install folder

home directories

- When instructions conflict

the following precedence applies:

- The agent must not make product

- **All state

memory

gemini.md

# gemini.md

- The agent must not introduce new technologies

frameworks

You value correctness

clarity

At the end of the simulation

the AI will analyze:

After your response

the AI will:

- AI may inject light humor in consequences (e.g.

“Your quick fix worked... until the coffee machine rebelled.”).

You will not be warned when latent risk is created

but watch for foreshadowing.

- Security

compliance

On each turn

you may take ONE meaningful action.

- Not all dependencies are visible

but subtle hints may appear in status updates.

Coding & Debugging

Engineers

incident responders

- 1.2 (2026-01-15) Introduced FUN ELEMENTS (light humor

stability points); set max turns to 10; added subtle hints and replayability via randomizable symptoms

Your job is to manage consequences

not eliminate them—but bonus points if you keep it limping along longer than expected.