Universal Tuya docsHomeSupported devicesRequest a deviceFAQChangelogArchitectureKnowledge baseCreditsSourcesContributing

Cross-research (P2936) — one full cross-check for every request

Every incoming item (needs-info sweep, open issues, forum reports, Homey Gmail diag / crash ids, CI reports) goes through the same pass: scripts/research/cross-research.js.

Sources crossed per item

SourceHowRole
Forum post + repliesDiscourse public JSON, cached 30 dcouple source
Same user's other postsuser_actions.json (30 max), cached 14 dhint only
GitHub issue + commentsgh issue view, cached 7 dcouple source
ScreenshotsOCR caches (forum-image-ocr.json, image-ocr.json) + tesseract (capped)couple source
Homey Gmail diags / crash mailsdata/diag-ids.json (ids only, built by diag-ids.js)couple source
Git historygit log --grephint only
Z2M, ZHA quirks, deCONZ DDF, Blakadder, tuya-local, Johan appshallow clones (fetch-refs.sh) → lib/ref-index.jsconfirmation + driver choice + model deduction
Web searchGoogle CSE only if XR_WEB=google-cse + key, cap XR_WEB_CAP (default 0)hint only
AIoff unless explicitly enabled within included quotahint only

Rules

Verdicts

Promotion / demotion (every run)

promoteDemote(): a provisional couple that a reference now confirms with a single target driver is moved there (and removed from the generic host); one that now conflicts (manufacturerName on a specific driver) is removed from the host and kept with status: demoted + reason. Nothing is silently deleted.

Workflows (job cross-research, composite .github/actions/cross-research)

forum-poll (forum), fleet-intelligent-enrich + auto-enrich-closed-loop (needs-info), gmail-diagnostics + fetch-diags (diags), auto-bot-issue-triage (issues), weekly-sovereign-loop (all). Pinned actions; caps on items, network calls (400), OCR (40), placements (10) and provisional (10) per run; cache restored between runs. Kill switch vars.XR_ENABLED=false, report-only vars.XR_APPLY=false. Commit only through safe-auto-commit.js; placements are reverted if the regression lock test fails.

Self-enrichment vectors (ranked value / risk)

#VectorValueRiskState
1Cross-research on every item, every workflowhighlowdone (this doc)
2Provisional tier + automatic promotion/demotionhighlow (generic host, gates)done
3Upstream incremental watch (index diff → upstream-delta.json, triggers promotions)highlowdone
4Ids from Homey diag / crash / mail / CI reports → diag-ids.json → cross-checkhighlowdone
5Confidence scoring on leads / couplesmediumnonedone
6Regression lock generated from each placement (data-driven test)mediumnonedone
7Runtime unknown DPs / frames (UnknownDPLogger) → diag → auto-mapping proposalhighmedium (wrong DP mapping)pending: proposals only, human review
8Per-device adaptive windows persisted (button debounce, report intervals)mediummedium (store writes)pending
9CI failure-pattern learning (athom-build-history)mediumlowpartial (existing history workflow)
10Optional AI layer within included quotaslow-mediummedium (hallucinated ids)pending, hints only, never a couple source

Credits: device identities cross-checked against Koenkk/zigbee-herdsman-converters (MIT), zigpy/zha-device-handlers (Apache-2.0), dresden-elektronik/deconz-rest-plugin DDF (BSD-3), blakadder/zigbee (MIT), make-all/tuya-local (MIT), JohanBendz/com.tuya.zigbee (GPL-3.0). Only fingerprints are used; all code is our own.

← Knowledge base index · source