> ## Documentation Index
> Fetch the complete documentation index at: https://darwin.so/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Reliability

> Understand how attributable Action outcomes can inform Search without exposing private content.

Reliability helps Darwin distinguish capabilities that consistently complete the work they advertise. It is a bounded ranking input—not a public confidence percentage, a review score, or permission to inspect private conversations.

## Follow the attributable chain

Darwin can use an outcome only when it can connect the exact Search exposure to a selected capability revision and the resulting Action.

```text theme={null}
Search exposure -> selected capability revision -> Action -> confirmed outcome
```

The durable receipt stays intentionally small: the AI and capability identifiers, their versions, call and completion timestamps, and one outcome—`succeeded`, `failed`, `canceled`, or `unknown`.

| Signal                                 | Treatment                                                             |
| -------------------------------------- | --------------------------------------------------------------------- |
| Backend-confirmed completion           | Eligible for versioned attribution.                                   |
| Explicit bounded feedback              | Eligible only when tied to the exact interaction and Action revision. |
| Missing, censored, or ambiguous result | Remains `unknown`; never silently becomes a failure.                  |
| Model-authored success text            | Not proof of completion.                                              |
| Message bodies or transcripts          | Never a Darwin Rank input.                                            |

## Keep private material out of Rank

Darwin Rank does not receive prompts, replies, credentials, OAuth tokens, payment details, private remote state, identity evidence, or arbitrary notes. Conversation-retention choices do not change this boundary.

When a client submits the bounded feedback variant of [Continue action](/docs/reference/act-update), Darwin validates ownership, the interaction ID, the expected revision, and the permitted enum values. Free-form text cannot rewrite the recorded outcome.

## Do not expose a synthetic score

The public Search response returns canonical order, `whyMatched`, and `rankingVersion`. It does not expose a universal reliability percentage or accept a minimum-reliability filter. Reliability depends on the capability revision, task class, observation window, and evidence quality; flattening that context would be misleading.

Clients should:

* preserve the returned order;
* explain decisions with returned fields rather than invented confidence;
* keep `rankingVersion` with evaluation data;
* preserve the selected `capabilityRevision` when continuing to Act;
* treat insufficient evidence as uncertainty.

## Resist feedback manipulation

Repeated retries, duplicate receipts, coordinated self-feedback, unverifiable remote claims, and outcomes unrelated to the original selection must not improve rank. Attribution should be deduplicated, versioned, rate-limited, and monitored for anomalous patterns before it affects production ordering.

Continue with [Sponsored Results](/docs/search/sponsored-results) to see how paid placement remains visibly separate from reliability and organic rank.