OSINT Hub
← Back to Science

Methods before claimsAn atomic, verifiable statement linked to evidence, a coverage envelope and separate human assessment dimensions. · draft

Negative results

A log of hypotheses that failed and honest abstentions.

reviewedlink-only

Capture-preflightPolicy validation of a URL, scheme, port, redirects and DNS observations before any network request is made. negative results

A control setA test set with expected answers that measures a reference implementation, not real-world quality. recording no-network blocks for SSRFA vulnerability class where an attacker causes a server to request an internal, local or otherwise forbidden address., DNS rebindingAn attack technique where one domain name changes between validation and connection to another, often internal, IP address. and unknown schemes.

EvidenceResearch material whose provenance, time, integrity state and link to an exact source version are preserved. / provenance

  • science/methods/capture-policy-preflight-v1.json
  • tests/test_capture_policy.py

Card limitations

  • This is a policy unit boundary, not a production egress test
  • A runtime owner must confirm the boundary separately
draftlocal-only

Search-abstentionA decision not to make a stronger conclusion when evidence is insufficient or conflicting, with the gap stated explicitly. examples

Examples of showing coverage and a safe next step instead of saying “no” when a result is missing.

Evidence / provenance

  • science/benchmarks/local-search-multilingual-v1.json
  • case:synthetic-media-geo

Card limitations

  • Abstention does not prove that data is absent
  • A negative resultA record that support was not found within a defined scope and period, or that a hypothesis failed, with its limits. is read with time and source boundaries
draftlocal-only

Non-comparable series

A synthetic correction path for when a schema or period change prevents interpreting a diffA deterministic comparison of two versions that shows an observed difference, not its cause..

Evidence / provenance

  • case:synthetic-public-diff
  • science/benchmarks/capture-policy-hostile-v1.json

Card limitations

  • An owner correction requires a new version and audit trail
  • A diff without a cause is not published

Science route tree

Method pages have separate states and never present draft as publication.

Method cardsDescribe a method's purpose, inputs, version and failure mode.draft BenchmarksSynthetic control sets test the reference implementation only.candidate Dataset cardsShow corpus provenance, rights, time window and limits.metadata-only ReproducibilityHelps repeat a safe scenario with a fixture, digest and versioned result.draft Negative resultsA log of hypotheses that failed and honest abstentions.draft Science governanceShows the owner, review and release gate for methods and data.candidate Science changelogKeeps dated semantic changes to methods and benchmarks.draft