v0.1.18 — the multi-provider cross-lab build

The multi-provider cross-lab build. Rung A — the opt-in, off-by-default independent second pass (Step 10) — becomes provider-agnostic. Every part was independently adjudicated by Fable 5 before landing (the same-lineage guard and the adapter loader adversarially). The adjudication layer stays opt-in and off by default, and its decision-flip benefit remains unproven — it is a rigour-and-defensibility check that sharpens a sound analysis, not a safety net, and it never flips the call.

What's new

• Predefined labs for rung A: OpenAI (default gpt-5.6-sol), Google Gemini (gemini-3.5-flash), and xAI Grok (grok-4.5), each via its own adapter, chosen with the crosslab_provider setting.
• "Other" mode: any OpenAI-compatible endpoint you configure (base URL + key + model), or an opt-in, hash-pinned, shape- and lineage-validated operator-written adapter file (default off; it runs your own code, so it is gated with a warning).
• Hard same-lineage guard: any Anthropic-lineage target on rung A is refused with no override — rungs B (Fable) and C (Opus panel) remain the honest same-lineage path.
• Switchable settings v3: prefs schema v2→v3, now 19 keys (adds provider, base-url, key-env-name, lineage, adapter-files, adapter-name); a shared prefs→env resolver with provider auto-inference and a provider/model-mismatch refusal.
• Monitor AMENDMENT-3: a per-provider split of the rung-A yield read-out.

Verification honesty

OpenAI is live-verified (a real smoke returned a clean parse). The Google Gemini and xAI Grok adapters are built to each lab's current published API (2026-07-09) but not live-smoked — they are live-verified only when you run the preflight smoke with that lab's own key. The keyless in-boundary rungs (B Fable / C Opus panel / D self-adversarial) are unchanged.

Install

Download insight-engine-0.1.18.plugin below and, in Cowork, Customize → upload from file.

v0.1.17 — the switchable-settings adjudication build

The switchable-settings adjudication build for the opt-in independent-adjudication layer (Step 10). Every part was independently adjudicated by Fable 5 before landing. The adjudication layer stays opt-in and off by default, and its decision-flip benefit remains unproven — it is a rigour-and-defensibility check that sharpens a sound analysis, not a safety net, and it never flips the call.

What's new

• 13-key switchable settings (schema v2, code-backed in prefs/): every adjudication behaviour is operator-switchable, standing and per-run; nothing is centrally locked. Ask-removing / egress-widening / block-reopen / breadth-raising changes require a deliberate confirmation.
• Privileged cross-lab gate changed from "never waivable" to default-blocked but operator-overridable by a deliberate typed act, logged to the local monitor (never the deliverable).
• Tagged cross-lab errors (CROSSLAB-BLOCKED / CROSSLAB-FAILED [...], HTTP-code classified) and a configurable multi-provider crosslab_model.
• Panel sizing — Opus panel N, cross-lab breadth M, runs-per-model — all as on-divergence offers, never automatic.
• Monitor AMENDMENT-2: a pre-registered per-class retirement rule for deciding, from real use, when adjudication has become redundant for a class of problem.
• The standing crosslab_model preference now retargets a run (CROSSLAB_MODEL remains the per-run override).

Install

• Cowork: download insight-engine-0.1.17.plugin attached below.
• Claude Code marketplace: the insight-engine plugin at v0.1.17.

v0.1.16: consecutive 1-12 step numbering (render/track split)

v0.1.16 — consecutive 1-12 step numbering

• Consecutive 1-12 step numbering. The analyse pipeline now uses a single, consecutive step scheme: triage (1) through decision brief (9), adjudication (10), render (11), track (12).
• Render and track are now separate steps. The former combined "follow-ons" step is split into Step 11 (render) and Step 12 (track).
• Adjudication relabelled Step 10. The old "L4.5" tag is replaced by "Step 10" across the adjudicate skill and its preferences (schema, prefs.py, README).
• Skill fixes. Restored the analyse and adjudicate skill files (recovered truncated tails) and corrected two stale internal cross-references in analyse (resilience check to Step 7; decision-brief items to 9d/9e).

Numbering, labelling, and consistency release — no change to pipeline behaviour.

insight-engine 0.1.15

v0.1.15: code-backed standing opt-out for adjudication

Makes the L4.5 adjudication standing opt-out mechanical instead of
model-remembered, closing the T-L45-001 B4 carry-forward.

New skills/adjudicate/prefs/:
- prefs.py: stdlib CLI — status / opt-out --scope crosslab|all /
  re-enable / should-offer --rung (exit 0 offer, 1 suppressed) /
  selftest. Persists to ~/.insight-engine/preferences.json (env
  override INSIGHT_ENGINE_PREFS), outside the plugin so it survives
  upgrades and is never committed. Fails safe to "offer" on a
  missing/corrupt file; no rung ever auto-runs.
- preferences.schema.json: draft-07 schema; scope enum none/crosslab/all.
- README.md: location, schema, scope semantics, exact commands.

Wiring:
- adjudicate/SKILL.md: gate now consults prefs.py should-offer before
  offering and persists via opt-out / re-enable.
- analyse/SKILL.md: Step 6.5 honours the persisted opt-out by code.

Packaging:
- plugin.json + marketplace.json bumped 0.1.14 -> 0.1.15.
- dist/insight-engine-0.1.15.plugin repackaged (22 files; clean-scanned
  for keys and real-case data; crosslab.py intact).

Verified: prefs selftest passes all transitions incl. corrupt-file
fail-safe; harness offline preflight and mock adjudicator still green.

insight-engine 0.1.14

v0.1.14

insight-engine 0.1.14 - opt-in adjudication layer

Insight Engine v0.1.7

Full Changelog: https://github.com/ComplexAR/Insight-Engine/commits/v0.1.7