PyOD 3.2.1 — od-expert skill correctness + demo redesign

Point release fixing correctness regressions shipped in the v3.2.0 od-expert skill, with a redesigned interactive demo and a pytest safety net to prevent this class of regression going forward. No breaking API changes.

What's fixed

• 17+ API reference errors in the v3.2.0 skill content: state.plan[...] → state.plans, state.scores → state.consensus['scores'], state.best_detector → state.analysis['best_detector'], and a phantom engine.start(X, y=labels) supervised path (there is no y parameter on ADEngine.start). Labelled data should use the classic XGBOD.fit(X, y) / predict(X) path; this is now what the skill and docs recommend.
• Invented profile keys removed. state.profile['estimated_contamination'] and state.profile['encoder'] were cited in v3.2.0 prose but do not exist on InvestigationState. The skill now tells the agent to observe state.analysis['consensus_analysis']['anomaly_ratio'] post-run instead.
• Hardcoded state.plans[:3] lists re-probed against a live ADEngine and corrected across references/tabular.md, references/time_series.md, references/graph.md, and references/text_image.md. The decision tables in those files are now explicitly labeled as expert heuristics, not predictions of engine.plan output; the agent is pointed at state.plans for the live selection.
• Cardio walkthrough reproducibility. references/workflow.md now seeds np.random.seed(42), excludes the trailing label column (X = df.values[:, :-1]), and reports the numbers that one-pass code actually produces (172/1831 flagged at the default contamination of 0.1, validation precision 85/172 at recall 85/176).

What's new

• pyod/test/test_skill_api_refs.py — a pytest safety net that walks ADEngine + InvestigationState via live dry runs (tabular / time series / text) and validates every state.X / state.X['a']['b'] / engine.X(...) reference in the skill content. Catches invalid keyword arguments through inspect.signature. Ships a synthetic negative test that fabricates all five regression shapes (bad attr, bad nested key, bad kwarg, missing method, missing prose attr) and asserts the scanner flags them.
• Redesigned interactive demo at examples/agentic_demo.html — now uses a diabetes screening dataset (examples/data/pima.csv, 768 patients, 8 features) with dark "od-expert decisions" callouts alongside the agent's turns showing modality triage, top-10 pitfall checks, the 11 adaptive escalation triggers, and the resulting plan. Two-column CSS grid with sticky callouts and overflow guards for narrow viewports. (Previously Pima Indians Diabetes; renamed to diabetes screening in user-visible text.)
• scripts/render_agentic_demo.py — Playwright headless-Chromium script that regenerates docs/figs/agentic-demo.png from the HTML demo source. Re-runnable any time the HTML changes so the readthedocs figure stays in sync.
• Expanded docs at docs/examples/agentic.rst with a new "What the skill encodes" section documenting the master decision tree, top-10 pitfalls, 11 escalation triggers, on-demand reference files, KB-derived detector list, and CI safety nets.

Review notes

• 4 rounds of /implement-review with Codex on this batch; 9 findings total (3 High + 4 Medium + 2 Low), all resolved.
• Highlight: the new safety net was designed around the exact regression shapes Codex's Round 1 review surfaced, plus Round 2's finding that the first version of the test missed nested dict-key chains and impossible call signatures.

Install

pip install --upgrade pyod
pyod install skill    # refresh ~/.claude/skills/od-expert/ with the fixed content
pyod info

v3.0.0 / v3.1.0 / v3.2.0 user code keeps working unchanged.