docs(knowledge): enqueue whats-new-opus-5 on custody grounds by kyle-sexton · Pull Request #1911 · melodic-software/claude-code-plugins

kyle-sexton · 2026-08-04T04:10:46Z

Summary

knowledge 0.10.21 — moves whats-new-opus-5 from "Deferred with trigger (not queued)" to the Models queue in the docpage-digest Anthropic docs profile, on custody grounds rather than the (unfired) fleet-lane trigger:

The playbooks Opus 5 model-adaptation chapter cites the page as sole authority for three shipped claims — thinking on by default; the 400 returned when thinking is disabled above effort high; and the live effort-level enumeration that establishes the upstream Opus 5 prompting guide's own ladder statement as truncated.
The models overview page carries none of those facts ("Adaptive thinking: Yes" is a capability flag, not a default-on statement; no 400 constraint; no ladder enumeration) — so the deferral's premise, "the overview covers it canonically", is false for exactly the facts already cited.
Scoped as a custody fact about this one page, not a decision to start a release-notes corpus: whats-new-sonnet-5 carries no such citations and stays deferred with its trigger unchanged.

The original trigger text is preserved verbatim in the changelog entry and commit body (the profile format keeps no history). The fleet-lane trigger itself was independently evaluated and has NOT fired (Opus 5 GA and the fleet pin both predate the deferral's own recording).

Test plan

Docs-only (profile entry move, changelog, version bump).
scripts/check-changelog-parity.sh --check, --check-bump origin/main, --check-order all pass; markdownlint clean; both plugin.json parse.
Independent fresh-context Fable verifier, 4 binary criteria — its own live fetches of both pages (confirming the three facts live on whats-new-opus-5 and absent from overview), citation-attribution check (including the corrected truncation attribution), entry-format and scope checks, changelog verbatim-trigger check, parity scripts re-run — 4/4 PASS, empty defect list.

No linked issue.
Doc-alignment loop, roster row 5 follow-up (independent-grounds enqueue; the row's own trigger evaluation stands as NOT FIRED). Related rows: feat(playbooks): carry the Opus 5 guide's correction-narration delta and re-verify its sources #1910 (row 3, whose chapter carries the citations needing custody).

🤖 Generated with Claude Code

https://claude.ai/code/session_019gaVX25Txd6GXdiu9HEH3X

The Anthropic docs profile deferred this page as "release notes for a model the models `overview` page already covers canonically; enqueue when Opus 5 enters or materially changes a fleet lane". That trigger has not fired. Custody moved it instead: the playbooks Opus 5 model-adaptation chapter cites the page as sole authority for three shipped claims — thinking on by default, the 400 returned when thinking is disabled above effort `high`, and the live effort-level enumeration that establishes the upstream Opus 5 prompting guide's own ladder statement as truncated — and `overview` carries none of them, so the deferral's premise is false for exactly the facts already in use. Scope is this one page. `whats-new-sonnet-5` has no such citations and keeps its identical trigger. Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

cursor · 2026-08-04T04:10:50Z

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

claude · 2026-08-04T04:11:13Z

Claude finished @kyle-sexton's task in 2m 41s —— View job

Review complete

Gather context (diff, changed files, CLAUDE.md/REVIEW.md guidelines)
Understand the request (code review of PR docs(knowledge): enqueue whats-new-opus-5 on custody grounds #1911)
Review anthropic-docs-profile.md change
Review CHANGELOG.md entry
Review plugin.json version bump
Post final review feedback

Scope note: this is the code-review lane (REVIEW.md). This PR is docs-only with no runtime/security surface, so there's nothing for the security lane to pick up, and no atomicity concern (three files change together in one commit; not related runtime state).

Findings

No correctness or convention issues found. Specifics I checked:

Custody claim is factually accurate. I cross-checked plugins/playbooks/reference/model-adaptation/opus-5.md (unchanged in this PR, but cited as the justification). It confirms whats-new-opus-5 is the only citation backing all three claims the profile entry lists:
- "Thinking is on by default... above [effort high] the API rejects... with a 400" (opus-5.md:111-113)
- "The guide's own ladder statement is TRUNCATED (verified against the live whats-new-opus-5 enumeration)" (opus-5.md:105-107)
So "sole authority for three shipped claims" in the new entry (anthropic-docs-profile.md:223-231) is not an unverified assertion — it holds against the current state of the cited file.
Verbatim-trigger preservation claim checks out. The original deferred-entry wording ("release notes for a model the models overview page already covers canonically; enqueue when Opus 5 enters or materially changes a fleet lane") is reproduced exactly in both the CHANGELOG entry (CHANGELOG.md:12-13) and the commit body, satisfying the PR's own claim that the profile format keeps no history.
No duplicate/orphaned entries. whats-new-opus-5 now appears exactly once (in the Models queue); whats-new-sonnet-5 is untouched and correctly left in "Deferred with trigger" with its own independent, unfired trigger — the scope-narrowing claim in the PR body matches the diff.
Version/changelog parity. plugin.json 0.10.20 → 0.10.21 matches the new ## [0.10.21] CHANGELOG header, consistent with this plugin's established pattern of patch-bumping doc-queue-only changes (e.g. docs(knowledge): repopulate the doc queue with the adopted enqueue roster #1895, docs(knowledge): admit anthropic.com/engineering and add cross-slice synthesis artifact target #1892).
Fresh-docs mandate applicability. Per CLAUDE.md's scoping rule, this file is a skill's context/prose body, not frontmatter/manifest/contract surface — so the mandate's live-WebFetch-and-cite requirement doesn't strictly apply here. The PR body's description of independent live verification (dated 2026-08-03) is a reasonable practice regardless.

Not verified in this pass: I could not execute scripts/check-changelog-parity.sh (--check, --check-bump origin/main, --check-order) or a JSON parse of plugin.json — shell execution beyond read-only git/grep/find was blocked by this session's tool-approval settings. Manual inspection (version/header parity, single-occurrence grep, valid-looking JSON diff) turned up nothing wrong, but if you want the scripts' own pass/fail signal independently confirmed, that needs a run with broader --allowedTools.

No blocking issues. This is ready to merge on correctness grounds.

claude · 2026-08-04T04:11:21Z

Claude finished @kyle-sexton's task in 50s —— View job

Security review

Fetch PR diff (origin/main...HEAD)
Review changed files for security issues
Report findings

Scope reviewed (per gh/git diff against origin/main, matching the PR's listed changed files):

plugins/knowledge/.claude-plugin/plugin.json — version bump only (0.10.20 → 0.10.21)
plugins/knowledge/CHANGELOG.md — new [0.10.21] prose entry
plugins/knowledge/skills/docpage-digest/context/anthropic-docs-profile.md — moves one URL bullet from the "deferred" section of the doc queue to the "Models queue" section, with rationale prose

Findings

None. This PR is a pure documentation/content change:

No code execution paths, scripts, hooks, or GitHub Actions workflows are touched.
No permissions: blocks, pull_request_target/workflow_run triggers, or run: blocks are touched — there is no CI/workflow surface in this diff at all.
No secrets, tokens, or credentials appear or are handled.
No new external inputs are introduced — the added content is a static URL bullet (https://platform.claude.com/docs/en/about-claude/models/whats-new-opus-5) already present elsewhere in the same file (it's a relocation from the "deferred" list, not a new external reference), plus prose describing custody/citation rationale for the knowledge skill's docpage-digest process.
No injection surfaces (command, SQL, path, template) exist in a markdown-content and JSON-version-field change.
No authorization/access-control logic is present or modified.

Nothing in this diff falls into the categories this review hunts for (logic flaws, auth gaps, injection, credential handling, dangerous Actions patterns, supply-chain/permission changes). No PR review submitted, as there are no findings to report.

github-actions · 2026-08-04T04:14:15Z

Claude has reviewed this PR 1 time. The lane skips further automatic reviews after 5; deleting this comment resets the count.

) ## Summary Refreshes `docs/PLUGIN-PHILOSOPHY.md`'s tier-to-model mapping, whose own recheck trigger — "a new Claude model family reaches GA, or the session default model changes" — fired both ways on 2026-07-24 (Opus 5 GA; the fleet session pin moved to `opus[1m]`). The table had been stale for eleven days, still naming Opus 4.8, which the live models overview now lists under Legacy models. | Tier | Old (2026-07-22) | New (2026-08-04) | |---|---|---| | Consequential verdict (session tier or above) | Fable 5 / Opus 4.8 | Opus 5 at the session tier; Fable 5 above it | | Mechanical prep, one tier down | Sonnet 5 | Sonnet 5 (re-verified) | | Bulk mechanical sweeps | Haiku 4.5 | Haiku 4.5 (re-verified) | Grounded in live fetches of code.claude.com/docs/en/model-config and platform.claude.com models overview at authoring time (and again by the verifier): `opus` resolves to Opus 5 on the Anthropic API; Fable 5 is positioned above Opus ("most capable", "not the default model"); Opus 4.8 is legacy. The ambiguous slash form is spelled out because the session tier and the rung above are now different models. Recheck trigger retained verbatim; no prices restated (pointer only); no Fable-specific row added — Fable's live differentiator is horizon, not work character, and a horizon tier would be a doctrine change, not a refresh. Also verified in passing: every agent-frontmatter `model:` value repo-wide is an alias (4 `opus`, 4 `sonnet`, 2 `inherit`) — the trigger's frontmatter re-audit finds nothing owed; and the flagged opus-4-8.md/ADR-0006/0007 staleness question was adjudicated as deliberate version-scoped architecture, not drift (Opus 4.8 remains upstream-available; the chapter's own header routes Opus 5 readers away). ## Test plan - Docs-only, single file, 14+/3−; markdownlint clean. - Independent fresh-context Fable verifier, 5 binary criteria — its own live fetches (quotes byte-exact, legacy placement confirmed), re-derivation soundness including an adversarial challenge to the no-Fable-row call and the session-tier semantics, style/scope, loop-lane §3 no-conflict check (the convention itself sanctions dated snapshot tables with recheck triggers; alias-only binds lane bodies), commit hygiene — **5/5 PASS, empty defect list**. ## Related - No linked issue. - Doc-alignment loop, roster row 5 follow-up (the table's own recheck trigger, surfaced during the row-5 trigger evaluation). Related: #1910 (row 3), #1911 (custody enqueue). 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_019gaVX25Txd6GXdiu9HEH3X --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

…ard and withdraw an unresolvable citation (#1913) ## Summary Doc-alignment roster row 4: **Claude Opus 5 System Card** (194-page PDF; live CDN copy byte-identical to the archived capture, SHA-256 re-verified by producer and verifier independently). **context-guard 0.4.5** — withdraws an unresolvable citation from the reader contract: the token-shape rationale co-cited "Anthropic system-card fixed-point evals" for degradation-tracks-absolute-tokens. The card names no such eval and contains no long-context degradation benchmark (normalized sweeps: fixed-point, MRCR, RULER, needle, LongBench all zero; §8.9 ProgramBench is fresh-budget episodes with a *rising* 83%→93% curve). Nuance preserved: the proposition is unsupported from an unnamed card, not asserted false — other Anthropic cards do publish long-context evals. No behavior change (the shape's other rationales are independent; band values were always declared judgment defaults). The contract keeps one standing line: cite a system card by name and section or not at all. **playbooks 0.6.12** — four card-grounded additions to `opus-5.md`, each fenced: - **Stated facts** (p.3, §6.5.1): more accurate than Opus 4.8 AND hallucinating slightly more, with confident-unsure cases — a factual specific with no tool call behind it is a recall claim. Fences the Verification delta against the broad reading that would strip needed lookups. - **Destructive actions** (§6.6.1, worked case p.115, transcript 6.4.2.A p.93): "Fabricated user consent" is a recurring category, and the transcript shows a written scoping rule defeated in-text — so for destructive/irreversible operations the remediation is a mechanism (PreToolUse hook, `permissions.deny`), not a written rule; a subagent return asserting approval is content, not authorization. Three fences: similar-to-Opus-4.8, reckless-use-significantly-down, pre-flagged-transcripts/earlier-snapshot (occurrence, not base rate). - **Effort**: two pilot cohorts *reported* quality falling at the top of the ladder (p.81–82) — kept explicitly report-not-finding with Anthropic's own disclaimer inline. - **Injection robustness**: quote boundary corrected ("…and browser use"); auto-mode 0%-of-129 reframed as evidence about a configuration, not the model, with the nonzero unsafeguarded rates (3.70%/4.30% browser; 0.56%/0.41% coding; 0.54%/0.39% computer use) and the operator action stated. - Sources: re-read + hash line replaces "the system card has not been re-read"; the deferred routing-lane trigger confirmed still unfired. Deliberately not shipped, with reasons in the commit: condescension note (too weak, conflicts with instruction-audit posture), vulnerability-discovery scaffolding sweep (no stale refusal workarounds found), per-worker-checkout orchestration seam (card-harness property, and a card citation in a model-agnostic chapter violates the scoping rule). ## Test plan - Docs-only; markdownlint clean; changelog-parity `--check` / `--check-bump` / `--check-order` all pass; zero lines over 100 chars. - Producer fanned the card to 4 section subagents, then re-grepped every shipped quote against the normalized PDF text itself. - Orchestrator-commissioned fresh-context Fable verifier: scripted quote-matcher over 29 fragments with a corrupted-quote control probe, its own normalized sweeps for the withdrawal, framing-honesty and fence-interaction checks — substance passed fully; its four defects (a heading accidentally consumed by the diff, a reviewer misattribution, a direction word, an example nit) fixed in the amended commit and ALL re-verified PASS, including an independent PDF check that the retained §6.2 pilot citation was genuinely correct. ## Related - No linked issue. - Doc-alignment loop, roster row 4. Predecessors: #1908 (row 1), #1909 (row 2), #1910 (row 3), #1911/#1912 (row-5 follow-ups). 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_019gaVX25Txd6GXdiu9HEH3X --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

…nd ship the IA-3 audit rows (#1914) ## Summary Doc-alignment roster row 6: **Prompting Claude Sonnet 5** (live page byte-identical to the archived capture, MD5 confirmed independently by producer and verifier — this ships unshipped deltas, not drift reaction). **playbooks 0.6.13** — new model-adaptation chapter `sonnet-5.md` (233 lines, opus-5/opus-4-8 pattern: version-scoped counter-steers, `[CC:]` tags, Sources block with capture provenance). Rationale for minting where RA-3 once said don't: ADR-0007 has since settled where per-model doctrine lives (closing RA-3's blocking premise), and the reader guaranteed to get no chapter — a Sonnet worker dispatched at low effort for mechanical prep, exactly where the guide places the under-thinking risk — was the one most likely to need it. Cross-references updated both halves (fable-5 meta-rule 3 gains the chapter; its no-chapter example narrows to Haiku; opus-4-8's preamble decoupled to generic routing so new chapters stop editing their predecessors). **claude-config 0.21.3 / criteria 1.11.0** — the IA-3 cluster dispositioned, all 11 candidates: - Shipped: **I8-e** (forced status cadence, `Model scope: sonnet-5` — deliberately scoped, not an unscoped I8-d promotion: the promotion gate needs two model guides *stating* the claim, and the Fable 5 guide's "Longer turns by default" prescribes timeout/streaming adjustments, never cadence removal — verified negative stamped on the row; the two rows state their no-co-fire relationship in their own words). **I17-c** (a fixed thinking budget prescribed where adaptive reasoning silently ignores or hard-rejects it — the finding is the missing model-AND-release gate, never the mention; fenced off I17-a's `=0` claim). - Already shipped, no work: qualitative severity bars (I8-b already cites this guide; `review`'s severity.md already carries decidable per-tier tests citing it — closing RA-2 / IA-1(1b) / IA-3's severity row as three IDs, one fix). - Dropped with evidence (6): verbosity suppression (page says *tune*, and designed succinctness surfaces exist), weak-tool-reach + thinking-off nudges (0 instances; fires on designed surfaces), unscoped-"apply" (not mechanically separable — the IA-2(d) failure mode again), sampling parameters (a source-code lint neither audit skill owns; zero instruction surfaces in-repo; deferred with trigger: first consumer repo with a direct Messages API / Agent SDK call site), coding-product addendum (guidance → chapter section), computer-use knob (no surfaces). ## Test plan - Docs-only; markdownlint (1,023 files) 0 errors; changelog parity `--check`/`--check-bump`/`--check-order` pass; `instruction-scan.test.sh` 46/46; `validate-plugins.sh` pass. - Producer-side: independent fresh-context reviewer (rationale withheld) — 19 findings all resolved, including defeating an unscoped I8-e promotion and an overclaim on `MAX_THINKING_TOKENS` (now carries the API-vs-third-party nuance, which model-config states verbatim). Producer's first quote-checker was itself found broken and rebuilt; final run 20/20. - Orchestrator-commissioned Fable verifier: its own 30-span quote extraction from the committed diff against five live pages (30/30), I8-e gate check against the live Fable guide, I17-c grounded against live env-vars, drop spot-checks, RA-2 closure confirmation, loss-free rebase proof — 7/8 clean; its three changelog-granularity defects fixed and re-verified PASS. - Rebased onto main after #1913; version stacks asserted: playbooks 0.6.13 > 0.6.12 > 0.6.11; claude-config 0.21.3 > 0.21.2; criteria 1.11.0. ## Related - No linked issue. - Doc-alignment loop, roster row 6 — last model-page row of the priority head. Predecessors: #1908 (row 1), #1909 (row 2), #1910 (row 3), #1913 (row 4), #1911/#1912 (row-5 follow-ups). 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_019gaVX25Txd6GXdiu9HEH3X --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

…odel facts (#1915) ## Summary Doc-alignment roster row 8: **Models overview** — the docpage-digest profile's canonical model-fact freshness source. First repo capture of the page (29,119 bytes, MD5 recorded); systematic ten-dimension sweep of every repo surface stating a fact this page owns. Sweep verdict: the repo is clean — pointer-not-copy holds everywhere except one surface. The single fix: `prompts/loops/loop-lane-prompts.md` justified two lane assignments with bare, undated page-owned facts (`opus` freshest-cutoff; `haiku` 200k/oldest-cutoff). Both were *true*; the defect was discipline. The rationales now carry the sanctioned dated-matrix shape already used by `docs/PLUGIN-PHILOSOPHY.md`'s tier table: figures upstream-owned and not restated, a dated resolution (2026-08-04), and a recheck trigger naming the derived-ordering failure mode (either comparison can flip while every underlying figure stays correct). Operational instructions unchanged. Also verified in passing: the profile's canonical-freshness claim is structurally true (current matrix + legacy accordion in one fetch); #1911's three enqueue negatives re-confirmed against the live page; all agent `model:` frontmatter still aliases; legacy placements and tier orderings consistent repo-wide. Component opportunity (a stored capture-and-diff freshness probe) declined: no upstream sync path for a hand-copied materialization, a one-consumer count, and the lightweight trigger shape already proved itself by producing #1912 — deferred with a trigger recorded on the roster row. ## Test plan - Single file, +11/−2, no plugin/changelog touched (parity script is plugin-scoped — confirmed by reading it); markdownlint clean. - Independent fresh-context Fable verifier: its own live fetch and scripted ordering checks (both derived orderings true), three sweep spot-checks (all holds), positional-wording check, `git merge-tree` conflict-free against current main, decline-soundness and scope-call review — **all PASS, empty defect list**. ## Related - No linked issue. - Doc-alignment loop, roster row 8. Predecessors: #1908–#1914 (rows 1–6 + row-5 follow-ups). 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_019gaVX25Txd6GXdiu9HEH3X Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

kyle-sexton mentioned this pull request Aug 4, 2026

docs(philosophy): refresh the tier-to-model mapping for Opus 5 GA #1912

Merged

kyle-sexton merged commit 87bc8d2 into main Aug 4, 2026
32 checks passed

kyle-sexton deleted the docs/enqueue-whats-new-opus-5 branch August 4, 2026 04:16

kyle-sexton mentioned this pull request Aug 4, 2026

feat(playbooks,context-guard): ground Opus 5 doctrine in the system card and withdraw an unresolvable citation #1913

Merged

kyle-sexton mentioned this pull request Aug 4, 2026

feat(playbooks,claude-config): mint the Sonnet 5 adaptation chapter and ship the IA-3 audit rows #1914

Merged

kyle-sexton mentioned this pull request Aug 4, 2026

docs(prompts): date the lane rationales that restate overview-owned model facts #1915

Merged

Navigation Menu

Uh oh!

docs(knowledge): enqueue whats-new-opus-5 on custody grounds - #1911

docs(knowledge): enqueue whats-new-opus-5 on custody grounds#1911
kyle-sexton merged 1 commit into
mainfrom
docs/enqueue-whats-new-opus-5

kyle-sexton commented Aug 4, 2026

Uh oh!

cursor Bot commented Aug 4, 2026

Uh oh!

claude Bot commented Aug 4, 2026 •

edited

Loading

Uh oh!

claude Bot commented Aug 4, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Aug 4, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

kyle-sexton commented Aug 4, 2026

Summary

Test plan

Related

Uh oh!

cursor Bot commented Aug 4, 2026

Uh oh!

claude Bot commented Aug 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review complete

Findings

Uh oh!

claude Bot commented Aug 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Security review

Findings

Uh oh!

github-actions Bot commented Aug 4, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

claude Bot commented Aug 4, 2026 •

edited

Loading

claude Bot commented Aug 4, 2026 •

edited

Loading