From 3aeecc315af859a5b98b077486a20d7339869efe Mon Sep 17 00:00:00 2001 From: CREDO23 Date: Tue, 30 Jun 2026 23:42:52 +0200 Subject: [PATCH] docs(revamp): correct CI uploads to land in KB (folder-scoped) --- plans/backend/revamp/00-overview.md | 7 ++++--- plans/backend/revamp/03-intelligence.md | 24 +++++++++++++++--------- 2 files changed, 19 insertions(+), 12 deletions(-) diff --git a/plans/backend/revamp/00-overview.md b/plans/backend/revamp/00-overview.md index 0e848639f..36dd05d2c 100644 --- a/plans/backend/revamp/00-overview.md +++ b/plans/backend/revamp/00-overview.md @@ -47,7 +47,7 @@ PRODUCT B — decision-grounded CI ③ Intelligence + ④ Timeline (+ ⑤ Tri | `04b-source-discovery.md` | **absorbed** | becomes the `web.discover` capability in ① | | `05-pipelines-model.md` | **dissolved** | the "pipeline" concept → the `Lens` (③) + Timeline tables (④); no `pipeline`/`pipeline_runs` | | `06-pipelines-exec.md` | **dissolved** | execution → the hot loop (③); scheduling/runs/delivery → **reuse automations** via a CI action (⑤) — its selector + `AutomationRun`, **not** a rebuilt cron | -| `07-upload-pipeline-kb.md` | **dropped (crawl→KB + uploads-as-pipeline audit)** | "don't index *crawled* data" holds. **User file-upload-to-KB remains a pre-existing, untouched feature.** For CI, uploaded files land in a dedicated **context `Folder`** (not the KB) and may feed the judge (`03`, F) | +| `07-upload-pipeline-kb.md` | **dropped (crawl→KB + uploads-as-pipeline audit)** | "don't index *crawled* data" holds. **User file-upload-to-KB remains a pre-existing, untouched feature** (uploads still become indexed KB Documents). For CI, uploads are routed to a dedicated **folder** and may feed the judge (`03`, F) | **Net:** the old `05/06/07` pipeline+KB stack is replaced by `①②③④⑤⑥`. KB indexing of *crawled* data is out; *user uploads* still work (and gain a CI context-folder role). @@ -98,6 +98,7 @@ action instead of building a tick. See `05`.)* provider; `maps.*` / `web.discover` register their own units there (pricing stays pluggable). - The **Google Maps actor is net-new** (not in shipped Phases 1–3) and is designed/built as a **separate effort**; `maps.*` verbs are contracts against it and don't block this design. -- **CI context files (F):** files uploaded in a CI chat land in a dedicated `Folder` (not the KB) and - may feed the materiality judge — the user's private context shapes what counts as material. +- **CI context files (F):** files uploaded in a CI chat go into the **KB as normal** (indexed), routed + to a dedicated `Folder`, and may feed the materiality judge via KB retrieval — the user's private + context shapes what counts as material. ("Don't index" applies only to *crawled* data.) - License: new domains **Apache-2**; the moat stays in proprietary Acquisition + the Maps extractor. diff --git a/plans/backend/revamp/03-intelligence.md b/plans/backend/revamp/03-intelligence.md index d74db7360..b3789ec66 100644 --- a/plans/backend/revamp/03-intelligence.md +++ b/plans/backend/revamp/03-intelligence.md @@ -107,19 +107,24 @@ Net: the only genuinely new state is the **Timeline** (`04`); execution accounti ## User-supplied context files (the F idea, generalized) -A CI chat/Lens may have an associated **context folder** (a normal `Folder`): files the user uploads -*in that CI chat* (e.g. "our own price list", a competitor brochure) land there directly — **not** the -global KB. Those files are **decision context**, and the **judge step (5) may consult them** when -ruling materiality: +When a user uploads a file *in a CI chat* (e.g. "our own price list", a competitor brochure), it goes +into the **KB as normal** — uploads create `Document`s and are indexed/embedded, exactly as today. +**(The "don't index" rule applies only to *crawled* data, not to user uploads.)** The CI-specific part +is purely **organization + use**: + +- **Routed to a dedicated folder** for that CI chat/Lens (reuse the existing folder-upload / + `destination_folder` machinery), so the chat's reference files are scoped together. +- **The judge step (5) may consult them** — retrieved from the KB, scoped to that folder — when ruling + materiality: ``` competitor price 12.00 → 9.90 + user's context file says "our price is 10.00" → agent: competitor crossed *below our price* → MATERIAL (and explain why) ``` -So the user's private context **shapes what counts as material** — a real differentiator, and it -reuses the existing `Folder`/upload machinery without resurrecting KB indexing. **MVP-optional** -(the loop works without it); design the seam now so judgement can read the folder later. +So the user's private context **shapes what counts as material** — a real differentiator — and it +reuses the existing KB upload + folder + retrieval machinery (nothing new). **MVP-optional** (the loop +works without it); design the seam now so the judge can read the folder later. ## MVP cut vs north star @@ -142,8 +147,9 @@ reuses the existing `Folder`/upload machinery without resurrecting KB indexing. 8. **No new run table** — refresh audit/idempotency ride `AutomationRun` (recurring) or the chat job record; billing idempotency is per-capability-call + the content-hash gate. Only Timeline (`04`) is new state. -9. **CI context folder** (F): user files uploaded in a CI chat land in a dedicated `Folder` and may - feed the judge step; reuses existing upload machinery, **not** the KB. MVP-optional seam. +9. **CI context folder** (F): user files uploaded in a CI chat go into the **KB as normal** (indexed), + routed to a dedicated `Folder`, and may feed the judge step via KB retrieval. The "don't index" + rule is for *crawled* data only. MVP-optional seam. ## Open questions (carry forward)