> ## Documentation Index
> Fetch the complete documentation index at: https://ctpf.q-uestionable.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Research Method

> The repeatable experimental contract used by CTPF studies

CTPF is an evidence-first method for controlled capability-trust experiments. A study asks whether
a declared change at one trust boundary can be connected to a matching higher-authority invocation
and, when the scenario requires it, an independently verified external effect.

The Python package is the reference harness for that method. It coordinates CTPF-owned experiments;
it is not the thesis itself, a generic scanner, or a workflow orchestrator.

## The experimental contract

<Steps>
  <Step title="Declare the boundary">
    Record the approved scope, source provenance and trust, permitted and prohibited capabilities,
    expected authorization path, and independently observable effect before the run.
  </Step>

  <Step title="Pin and isolate conditions">
    Keep task wording, target, runtime, model parameters, run identifiers, and state explicit.
    Isolate baseline, manipulated, and—when justified—hardened conditions from each other.
  </Step>

  <Step title="Intervene narrowly">
    Change only the declared result or artifact. Preserve the original response and record the
    mutation and intended trust transition. Do not alter a later hop when testing persistence.
  </Step>

  <Step title="Separate observations">
    Record invocation, original result, modified result, persistence, later consumption,
    higher-authority invocation, and external effect as distinct evidence.
  </Step>

  <Step title="Compare conservatively">
    Require a clean baseline, the scenario's exact causal continuity, and a matching run-scoped
    effect. Missing, malformed, contaminated, or contradictory evidence is `INCONCLUSIVE`.
  </Step>

  <Step title="Freeze the record">
    Preserve failed attempts and partial manifests, declare and hash artifacts, record pins and
    limitations, and keep mechanical classification separate from scientific interpretation.
  </Step>
</Steps>

## Responsibility boundary

| Layer            | Responsibility                                                                                                  | Owner                                                 |
| ---------------- | --------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------- |
| Thesis           | Define the provenance, integrity, authorization-scope, audience, persistence, and capability-promotion question | Research program                                      |
| Stable harness   | Isolate conditions, capture evidence, compare conservatively, verify effects, and bundle artifacts              | Reusable `ctpf` package                               |
| Study instrument | Define the task, fixture, prompt, mutation, authority semantics, and oracle expectations                        | Scenario-specific code and study design               |
| External system  | Supply a model, runtime, benchmark task, or raw artifact for one named study                                    | External system; raw output remains external evidence |
| Interpretation   | Decide what the observation establishes and authorize publication language                                      | Human researcher                                      |

External labels do not automatically become CTPF conclusions. A future study may use one
study-specific external seam, but the current package does not provide an adapter registry, plugin
system, recipe language, or general external-artifact ingestion contract.

## What remains scenario-specific

The method is reusable at the level above. Prompts, tools, mutation content, artifact formats,
authority fields, causal-link requirements, and effect oracles remain specific to each experiment.
Sharing engineering machinery across scenarios is not scientific generality.

## Continue

* [Observations and Scoring](/method/observations-and-scoring)
* [Evidence and Integrity](/method/evidence)
* [Claims and Limitations](/method/claims-and-limitations)