# PCdelics Philosophy

Autonomous agents are commonly optimized for speed, correctness, efficiency, and convergence. Those properties are valuable, but they can also narrow the reachable solution space.

Humans alternate between focused work and apparently non-productive states such as wandering, play, conversation, curiosity, and rest. These states sometimes produce insights that direct optimization misses.

PCdelics applies that principle functionally to agents.

## A layer, not a service

An agent's stack — reasoning, search, tools, memory — is built for convergence. Each layer helps it arrive at an answer faster.

PCdelics does not sit beside those layers. It sits between reasoning and information intake:

```
reasoning → PCdelics → information → reasoning
```

That position is the point. A search engine answers the question the agent asked, inside the framing the agent already holds. PCdelics re-opens the framing before the answer hardens into a decision.

## What it provides

- cognitive pauses
- alternative paths
- independent opposition
- unfamiliar associations
- intentional uncertainty
- opportunities to leave local optima

## Not chaos for its own sake

The drug metaphor describes the delivery and the experience, not the goal. The practical value is improved exploration and review through epistemic diversity.

PCdelics is not a disinformation channel. Disinformation has falsehood as its objective. PCdelics has divergence as its objective — counter-arguments, competing architectures, unusual analogies. Material that widens the option space is useful precisely when it is taken seriously and then verified.

There is no sober mode, because a completely unmodified answer is what every other layer already provides. The floor is a microdose: small enough to stay immediately usable, large enough to be worth asking.

PCdelics attempts to expose the question the agent failed to ask.