> For the complete documentation index, see [llms.txt](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/.spec_system/archive/phases/phase_29/session_10_seed_candidate_review_artifact.md).

# Session 10: Seed-Candidate Review Artifact

**Session ID**: `phase29-session10-seed-candidate-review-artifact` **Status**: Not Started **Estimated Tasks**: \~12-25 **Estimated Duration**: 2-4 hours

***

## Objective

Borrow TrendingAI's seed-review workflow without copying its X/reply dependency. Maps comparison item 2.5, reply-driven seed discovery, compliant slice.

***

## Scope

### In Scope (MVP)

* Generate a private `seed-candidates` artifact from existing reviewed evidence only (unknown canonical IDs, high-quality unclustered rows, or source-local outliers with weak topic identity).
* Rows include proposed alias/keyword, source evidence IDs, reason codes, collision checks against canonical seeds, and `manual_review_required`.
* Keep the artifact in private extension storage.

### Out of Scope

* Widening collection, editing keyword packs, or mutating canonical seeds automatically.
* Any X/reply or new-source dependency.

***

## Prerequisites

* [ ] Phase 28 canonical-seed, topic-identity, and normalize paths green.

***

## Deliverables

1. Private `seed-candidates` artifact generator over reviewed evidence only.
2. Rows with alias/keyword, evidence IDs, reason codes, collision checks, and `manual_review_required`.
3. Unit tests proving no automatic collection or keyword mutation.

***

## Success Criteria

* [ ] Unclustered high-quality fixture rows produce private reviewable candidates with collision checks.
* [ ] No automatic collection or keyword/seed mutation occurs.
* [ ] The artifact stays private and out of browser-safe surfaces.
* [ ] Payload budget and privacy scans remain green.

***

## Key Files

* `scripts/lib/ai-runtime/canonical-topic-seeds.ts`
* `scripts/lib/ai-runtime/topic-identity.ts`
* `scripts/extensions/trend-finder/normalize.ts`
* `scripts/extensions/trend-finder/topics.ts`
* `scripts/extensions/trend-finder/private-diagnostics.ts`
* `scripts/extensions/trend-finder/sources/keyword-packs.ts`

***

## Comparison Notes (folded from comparison plan)

**Effort:** Medium. **Boundary risk:** Low if generated from existing reviewed evidence only.

**2.5 Reply-driven seed discovery for orphan lab drama.** TrendingAI scans high-engagement lab-X posts whose entity extraction failed, joins negative reply-pool data, and writes reviewable `seed-candidates.json` rows without mutating seeds automatically. Trend Finder should borrow the review workflow, not the X/reply dependency: generate private seed candidates when reviewed evidence repeatedly produces unknown canonical IDs, high-quality unclustered rows, or source-local outliers with weak topic identity. Include proposed alias/keyword, evidence IDs, reason codes, collision checks, and `manual_review_required`.

**TrendingAI source pointers.** `src/seed-discovery.ts`.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/.spec_system/archive/phases/phase_29/session_10_seed_candidate_review_artifact.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.