Intel
Time-stamped developments on Projects, graded by evidence so the board reads at a glance
Intel captures discrete developments on Projects: a launch, a listing, an exploit, a filing. One intel is not a single post. It is an evolving record that accumulates observations as more sources pick the story up, and it keeps the full trail of how the story changed.
Clusters show which account archetypes have picked up the development.
How to read a gem
Every intel renders as a gem; four properties carry the evidence.
| Property | Reading | |
|---|---|---|
| Color | Sentiment: green positive, red negative, gray neutral or not yet classified | |
| Cut | The evidence grade, from rough shard to fully cut stone | |
| Glow | The parent Project's score is moving right now | |
| Sparkle | Observed in the last hour |
The cut ladder
| Cut | Meaning | |
|---|---|---|
| Chipped | A single sighting with no source attached | |
| Flawed | Echoed by more posts or clusters, still unsourced | |
| Cut | Carries at least one externally verified citation | |
| Polished | An official Project account is in the record | |
| Flawless | Official, corroborated, and broad across clusters |
Cut grades the shape of the evidence, never the truth of the claim. Thresholds keep the top grade rare: well under one percent of intel is Flawless.
For developers
Intel responses include the derived cut grade, and list endpoints accept a
cuts filter to keep only chosen grades. See the
REST API Reference for the full intel contract.