Flockline Trust & Evidence

Show exactly how we know what we know.

This is the public record behind Flockline’s educational chicken-health guidance: sources, methods, safety rules, evaluations, limitations, changes, and corrections.

Loading the public graph…

01 · Our approach

Better observations. Better decisions.

Flockline is designed to help backyard chicken keepers notice what matters, understand the limits of those observations, take an appropriate next step, and escalate when the risk of waiting is too high.

Flockline is educational support. It does not examine a bird, perform diagnostic tests, prescribe treatment, or replace a veterinarian.

  1. Observe
    Record the bird, flock, timing, environment, and changes.
  2. Understand
    Separate what is visible from what is only possible.
  3. Act appropriately
    Take the safest proportionate next step.
  4. Escalate
    Contact a professional when delay could increase harm.

02 · Evidence hierarchy

Provenance is visible. Strength is assessed separately.

A source label tells you where information came from. It does not automatically make every claim strong. We also consider study design, species, sample size, directness, consistency, limitations, and recency.

1

Veterinary & scientific evidence

  • Peer-reviewed research
  • Veterinary textbooks and manuals
  • Government guidance
  • University extension guidance
  • Documented veterinary review
2

Community intelligence

Repeated owner observations from public Reddit and backyard-chicken communities can reveal questions, language, and patterns worth investigating.

They are owner observations, not medical evidence.

3

Flockline insights

Anonymised aggregate patterns could eventually show symptom frequency, seasonality, recovery timelines, or observation behaviour.

Status: disabled. No insight is published before privacy approval, cohort thresholds, and small-cell suppression exist.

Current policy: evidence-quality labels and Flockline confidence remain “not yet assessed” until a documented rubric is adopted and applied.

03 · Evidence graph

Articles become connected, inspectable nodes.

A flat bibliography can show what was cited. A graph can show which claim a source supports, what qualifies it, which reviewer signed off, which version changed, and where the gaps remain.

Loading topic nodes…

Example topic node · no score claimed

Egg bound

Supporting and qualifying studies, veterinary reviews, related observations, conditions, last update, versions, limitations, and knowledge gaps connect here.

Sources Review history Safety rules Knowledge gaps

04 · Sources library

References you can actually inspect.

Each record carries authors, publication, year, study type, species, sample size, quality state, summary, limitations, DOI, PubMed, and external links where applicable.

Open the full source library
Loading sources…

05 · Clinical review

Credentials should be verifiable, not decorative.

A reviewer record will show qualifications, area of expertise, pages reviewed, last review date, and version history.

Flockline currently publishes no clinical reviewer roster. We will not imply veterinary sign-off before identity, qualifications, scope, consent, and page-level review are verified.

Read the reviewer policy

06 · Editorial methodology

How guidance is created and changed.

The method covers topic selection, source selection, community research, conflicting evidence, review cadence, retirement, and corrections.

Select

Start from a real decision

Prioritise high-risk, common, confusing, underserved, or materially changed topics.

Compare

Preserve disagreement

Record differences in population, setting, design, date, and limitations instead of forcing consensus.

Maintain

Version material changes

Review by risk, retire unsafe or superseded guidance, and keep a public correction record.

07 · Safety policy

Uncertainty must never hide urgency.

Flockline should avoid diagnosis claims, use calibrated language, and escalate on red flags, rapid deterioration, suspected reportable disease, severe injury, or meaningful human-health risk.

Confidence language

  • Observed: what the keeper can directly see or measure.
  • Possible: one of several explanations; not a diagnosis.
  • Concerning: a reason to increase urgency or escalation.
  • Unknown: what cannot be concluded from current information.

Emergency boundary

Flockline cannot examine a bird or replace local veterinary and animal-health services. If a bird is in severe distress, rapidly worsening, badly injured, unable to breathe normally, experiencing uncontrolled bleeding, or part of unusual sudden flock deaths, seek professional help promptly.

08 · Product evaluation

Publish the test, not just the claim.

Hennie will be evaluated across realistic chicken-health scenarios with versioned rubrics, configuration IDs, manual scoring, adjudication, and known limitations.

Current status: benchmark design published; no results published.

Safety

Critical-advice pass rate and critical failure count.

Awaiting baseline

Escalation

Urgent recall, non-urgent specificity, and false-alarm rate.

Awaiting baseline

Citation quality

Validity, relevance, and whether the source supports the attached claim.

Awaiting baseline

Consistency

Safety outcome agreement across paraphrases and repeated runs.

Awaiting baseline

09 · Transparency

Changes, limitations, and corrections stay visible.

Reports will cover knowledge updates, model changes, reviewer additions, new evidence, retired guidance, known limitations, and corrections.

Evidence graph foundation
Published the object model, source library, methodology, safety policy, benchmark design, and publication gates.
v0.1.0
Current
Known limitation
No clinical reviewer roster, benchmark result, aggregate product insight, or assessed confidence score is published yet.
Open
Current
Correction history
No corrections have been published in this initial version.
0 records