Evidence and field reports

Feel the use. Show the run.

AIppocampus should be judged by source-backed runs, bounded claims, and reports that say exactly what was tested.

Magic moments come before the evidence wall.

A useful field report does not have to pretend to be a benchmark. It should show the moment where continuity changed the work, then name the boundary of the claim.

Source-backed

The report points to a run, source snippet, benchmark, or reproducible path.

Claim-bounded

The report says what it proves, what it only suggests, and what still needs deeper testing.

Public-safe

Private rollouts, local paths, credentials, and private conversation text stay out of public output.

What goes where.

SurfaceUse it forBoundary
Official evidence mapBenchmarks, smoke runs, readiness claims, and known gaps.Repository docs remain the source of truth.
Community reportsConcrete examples from real users and downstream setups.Signals, not automatic product claims.
DiscussionsQuestions, ideas, design review, and developing interpretation.Keep exploratory threads in the open conversation.

Use this report shape.

Field report template
## What I tested

Describe the AIppocampus scenario, command, hook, or workflow you tried.

## What continuity changed

Explain what became easier because source-backed continuity was available.

## Source boundary

State whether the result came from exact source, bounded evidence, a route, or a general impression.

## Environment

List the agent host, operating system, AIppocampus version or commit, and any relevant provider.

## Gaps or surprises

Name anything that failed, felt confusing, or needs more evidence before becoming a project claim.

Official evidence has a first-stop map.

Public site pages summarize. Repository docs keep the magic moments, claim ladder, provenance ledger, benchmark evidence, and readiness boundaries.