A fourth explicit verdict, "AI slop" (Vote::Slop, label `slop`), joins the
rating widget on the web, the EPUB footer links, the /r/ confirmation page,
the CLI, imports and the dashboard filters. It counts as a full negative
(curation.feedback.slop_value, -1.0) in the neighbour and affinity signals.
Beyond that, each run loads the authors whose current verdict is slop, with
no lookback, and multiplies the preliminary blend and the utility of every
candidate by that author by 1 - curation.ranking.slop_author_penalty (0.75),
so they sink before triage. The flag is recorded in signals_json, shown by
`explain` and the dashboard signals table, and the confirmation names the
author (or says no author is known, in which case only the rating applies).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VXYGBPoHZSDSfE5WcJ9bvj
The article dashboard page gains a "Nearest articles (any)" table: the
ten closest stored embeddings by cosine, regardless of rating or run,
via a brute-force scan of article_embeddings. The deep-assessment format
facet grows from 5 to 14 values (code_repository, documentation_reference,
tool_or_product_page, discussion_thread, paper_or_report,
interview_or_transcript, video_or_podcast, fiction_or_humor, other) and
the triage kind from 10 to 16 (repo, docs, discussion, paper, media,
fiction), so a GitHub repository is no longer forced into analysis_essay.
Both prompt versions bump to 2.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QVPagF6jfDv78CC5Jv2wp4
DeepSeek's content filter rejects a whole request (400 "Content Exists
Risk") when one article trips it, which cost the other articles in the
batch their assessment and retried them every run. New curate/batch.rs
runs both stages through a bisecting runner: a rejected batch is split
until the offending article is isolated, that article is retried once on
the editor provider when it is a different one, and a still-rejected
article is recorded as a provider_rejected assessment row so it is not
retried for assessment_reuse_days. Cache reuse accepts rows from either
configured model. Each stage logs reused/requested/rejected counts, the
curation: line shows rejections when non-zero, explain prints them, and
the llm_assess span reports the deep-set size.
Implemented by a Claude agent from an orchestrator brief; verified
fmt/clippy(-W dead_code)/test green (354 lib tests).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A1rCLQeKBgnBo3oTgHuTMe
[llm] assigns the bulk and editor roles by name over a [providers.*]
registry (kind = openai | anthropic, per-provider model, effort, daily
ceiling and price table); DeepseekBackend becomes OpenAiCompatibleBackend
(reasoning_effort passthrough), AnthropicBackend builds from the same
ProviderConfig, meters and provider_costs are keyed by provider name.
Gemini 3.8 Flash is declared via Google's OpenAI-compatible endpoint so
switching the editor is one line (or DAILY_EPUB_LLM__EDITOR=gemini for an
A/B dry run). Stale [deepseek]/[anthropic] tables, the top-level
max_daily_usd and the old key env vars fail loudly.
daily-epub config check validates and prints the resolved roles, models,
key presence and paths without opening the database.
docs/runbooks/curation-v2-migration.md walks the server upgrade from v1.
Registry implemented by a Claude agent from an orchestrator brief;
verified fmt/clippy(-W dead_code)/test green.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A1rCLQeKBgnBo3oTgHuTMe
assess.rs replaces score.rs (DEEP_INSTRUCTIONS, representative sample
with [BEGINNING]/[MIDDLE]/[END], facets, cached deep rows with --rescore
bypass), rank.rs adds the utility blend over present signals with gate
ramps and the leader-clustered shortlist (cap 2 → 3 → uncapped, protected
top-N, exploration reserve), editor.rs replaces select.rs with the §13
rendering and utility-ordered fallbacks. ScoredArticle is gone; Candidate
is the only flow type. deep_batch_size replaces score_batch_size.
Started by Codex (cut off by its usage limit mid-verification) and
finished by a Claude agent from docs/plans/curation-v2-briefs/step5.md;
reviewed against plan §12–§13.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A1rCLQeKBgnBo3oTgHuTMe
thalladaandClaude Fable 5.1