> 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_01_editorial_quick_wins.md). # Session 01: Editorial Quick Wins **Session ID**: `phase29-session01-editorial-quick-wins` **Status**: Not Started **Estimated Tasks**: \~12-25 **Estimated Duration**: 2-4 hours *** ## Objective Raise editorial quality of creator-facing copy at near-zero cost, fold the reviewed engagement-farming hype-pattern set into existing noise scoring, and document the broader-social-reach non-goal. Maps comparison items 1.6, 1.5, and the 2.4 non-goal. *** ## Scope ### In Scope (MVP) * Append an ANTI-AI-TROPE block to `TREND_ANALYST_SYSTEM_INSTRUCTIONS` / `TREND_ANALYST_TASK_PROMPT` covering forbidden openers, hedges, hype terms, rhetorical reflexes, em-dash-as-soft-pause, and a preference for mechanism verbs over vague verbs. * Mirror the voice discipline into deterministic fallback copy templates. * Add a banned-phrase scan to the same analyst validator path that already rejects invented IDs, URLs, dates, source names, and metrics. * Add a reviewed creator-hype pattern set near `contentPatternRegexes` in `topic-quality.ts` (e.g. "save this thread", "DM me", "top 0.1%", numbered lists of six or more, hashtag/CTA closers) and feed the existing `topicNoiseDownrank`. * Add a known-coverage-gap paragraph documenting X/TikTok/Instagram/Bluesky as a deliberate non-goal under current ToS/PII posture. ### Out of Scope * Any new browser-visible field (hype detection feeds existing downrank only). * Any new source or collector for social platforms (2.4 stays a non-goal). * Reception polarity (Session 03) and attention pattern (Session 02). *** ## Prerequisites * [ ] Phase 28 completed; analyst validator and `topicNoiseDownrank` paths green. *** ## Deliverables 1. ANTI-AI-TROPE prompt block plus mirrored fallback-copy discipline. 2. Banned-phrase validator pass with fixtures for rejected and accepted copy. 3. Reviewed creator-hype pattern set wired into topic-quality noise scoring. 4. Documented social-reach non-goal paragraph in source docs. *** ## Success Criteria * [ ] A banned-phrase fixture fails analyst validation; legitimate copy passes. * [ ] A known engagement-farm title is downranked in topic-quality tests. * [ ] The 2.4 social-coverage gap is documented as a deliberate non-goal. * [ ] Deterministic fallback copy still renders; payload and privacy scans green. *** ## Key Files * `scripts/lib/ai-runtime/trend-analyst.ts` * `scripts/lib/ai-runtime/topic-quality.ts` * `scripts/extensions/trend-finder/sources/source-quality.ts` * `scripts/lib/ai-runtime/__tests__/trend-analyst.test.ts` * `scripts/lib/ai-runtime/__tests__/topic-quality.test.ts` * `docs/extensions/trend-finder-sources.md` *** ## Comparison Notes (folded from comparison plan) **Effort:** Low. **Boundary risk:** None for 1.5/1.6; Prohibitive under current posture for 2.4 (non-goal, not a build). **1.6 Anti-AI-trope copy discipline.** TrendingAI prompts include a strict ANTI-AI-TROPE block covering forbidden openers, hedges, hype terms, rhetorical reflexes, em-dash-as-soft-pause, and a preference for mechanism verbs such as "released" or "replaced" over vague verbs like "shift". Trend Finder's angle-pack copy and `whyNow` text have no equivalent voice guard. Add the block to `TREND_ANALYST_SYSTEM_INSTRUCTIONS` and/or `TREND_ANALYST_TASK_PROMPT`, mirror it in deterministic fallback templates, and add a banned-phrase pass to the analyst validator. **1.5 Tutorial-hype / engagement-farming noise detection.** TrendingAI uses a deterministic regex set for patterns like "save this thread", "DM me", "top 0.1%", "the only X you need", numbered lists of six or more items, hashtags, and CTA closers, then omits engagement-farming content even when it is technically accurate. Trend Finder already has `topicNoiseRisk`, `topicQualityGate`, and `generic-or-vague-topic`. Adopt the reviewed pattern set into topic/evidence quality scoring and feed existing `topicNoiseDownrank`; no new browser field is required. **2.4 Broader social reach (X/TikTok/Instagram/Bluesky).** TrendingAI harvests creator-native social platforms and reply threads. Trend Finder's source posture deliberately treats X/Twitter as a non-goal under current API/terms posture, Bluesky as deferred pending PII review, and TikTok/Instagram as outside the current candidate list. Keep this as a known coverage gap, not a collector task. The compliant creator-role slice remains reviewed YouTube via Apify. **TrendingAI source pointers.** Anti-AI-trope: every file in `prompts/` with the ANTI-AI-TROPE block. Tutorial-hype: `src/feature-sentiment.ts` (`TUTORIAL_HYPE_PATTERNS`), `scripts/classify-one-to-watch.ts` (`HUSTLER_FORMAT_MARKERS`). --- # 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_01_editorial_quick_wins.md?ask=&goal= ``` `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.