opencode-agent-skill 9.0.0 → 11.0.0

package/CHANGELOG.md

@@ -6,6 +6,122 @@ The project follows Semantic Versioning.
 
 ## [Unreleased]
 
+## [11.0.0] - 2026-09-22
+
+### Released
+- Promoted V11 perception-aware adaptive execution to stable after the full local CI gate passed on Windows with 234 tests total, 232 passed, 0 failed and 2 platform-specific skips.
+- Stable npm installs use the default `latest` dist-tag, so users install with `npm install -g opencode-agent-skill`.
+- Includes content-addressed Evidence Store, adaptive EvidenceBudget/context externalization, stable-prefix prompt telemetry, capability-aware model routing, visual geometry receipts, deterministic PNG diff/crop, responsive/design-token inspection, optional Playwright browser inspection, cost-aware dynamic workflow scheduling, 48 focused skills and 12 subagents.
+
+### Verified
+- Syntax: 178 JavaScript modules.
+- Catalog: 48 skills, 11 commands and 12 subagents.
+- Router: 129 cases, 294/294 required routes and 61/61 negative guards.
+- V11 contracts: 13 tasks across 6 categories and 15 required runtime files.
+- Live/long/polyglot fixture validation: 20 / 5 / 8 tasks.
+- npm pack dry-run, packed-install smoke and plain one-command install/resource auto-sync smoke: PASS.
+
+### Changed
+- Package version is 11.0.0.
+- V11 becomes the stable npm release line; V10 remains in Git history as the previous stable release.
+
+
+### V11 dev.2
+- Added project-local, fail-closed Playwright browser inspection that returns bounded semantic elements, bounding boxes, computed visual properties and a screenshot path while treating page content as untrusted evidence.
+- Exposed browser inspection through CLI and the OpenCode V2 router without making Playwright a required package dependency.
+- Fixed duplicate `ui_layout` router tool registration and aligned durable context packs with V11 context schema version 6.
+- Extended V11 validation and regression coverage for the browser runtime.
+- Package development version is 11.0.0-dev.2.
+
+## [11.0.0-dev.1] - 2026-09-22
+
+### Added
+- Adaptive context engine that externalizes oversized source/test/reference excerpts into content-addressed evidence pointers while keeping bounded inline previews.
+- Deterministic UI layout and design-token tools exposed to the OpenCode V2 runtime.
+- Cost- and modality-aware workflow scheduling with inline thresholds, deterministic-first waves, independent LLM/vision concurrency and bounded wave cost.
+- Repeated-stable prompt ratio telemetry and an optional fail-closed ablation gate for repeated input efficiency.
+- Stronger V11 contract coverage for adaptive context, UI inspection and evidence externalization.
+
+### Changed
+- V11 router metadata now reports version 11.
+- Task context packs transport large contextual evidence through `evidence:sha256` pointers and report externalized byte/ref counts.
+- Capability-aware model routing fails closed when an enabled configured model set cannot satisfy required capabilities such as vision/browser.
+- Package development version is 11.0.0-dev.1.
+
+### Fixed
+- V11 eval-report regression tests now account for expanded adaptive telemetry coverage instead of using the old V10-only telemetry shape.
+
+## [11.0.0-dev.0] - 2026-09-22
+
+### Added
+- Content-addressed Evidence Store with bounded retrieval, deduplication and garbage collection.
+- Adaptive Evidence Budget planning and context-manifest integration for evidence-efficient execution.
+- Prompt stable-prefix/cache telemetry for repeated-input measurement.
+- Capability registry and capability-aware model selection for coding, reasoning, tools, vision, browser, filesystem and long-context needs.
+- Visual specification, geometry receipts, responsive viewport matrix, dependency-free PNG decode/diff/crop and bounded visual repair planning.
+- Browser QA adapter with CLI-first verification planning, targeted semantic evidence and explicit untrusted-page security boundaries.
+- Cost-aware dynamic workflow scheduler separating deterministic work from LLM/vision work.
+- Skill-quality linting for entrypoint size, metadata and routing-description collision detection.
+- Nine V11 skills: visual-fidelity, browser-qa, design-source, responsive-verification, component-visual-testing, browser-security, skill-authoring, skill-evaluation and dynamic-workflow.
+- Two V11 subagents: visual-verifier and merge-arbiter.
+- Optional Hermes sidecar workflow contract with evidence-pointer transport.
+- V11 Control Center evidence-store/runtime telemetry and V11-specific tests/eval routing cases.
+
+### Changed
+- Model policy schema supports per-model capability metadata plus cost, latency and quality hints.
+- Adaptive model resolution can select configured models by required task capabilities.
+- OpenCode V2 router recognizes visual/browser/design/skill-workflow intents and keeps FAST routing selective.
+- Package version is 11.0.0-dev.0 while npm latest remains V10 stable until V11 release gates pass.
+
+## [10.0.0] - 2026-09-22
+
+### Released
+- Promoted V10 RC.2 to stable after the local full gate passed with 182 tests passing, 0 failing, 2 platform-specific skips, plus package and install smoke validation.
+- Stable npm installs use the default `latest` dist-tag, so users install with `npm install -g opencode-agent-skill`.
+- Includes adaptive context/routing, weak-model recovery, no-progress watchdog, duplicate/loop guards, bounded exploration output, durable compaction checkpoints, provider recovery and retryable lease recovery.
+
+### Changed
+- Package version is 10.0.0.
+- V10 stable keeps FAST at 8k, STANDARD at 20k and DEEP at 48k while preserving configured executor capability and correctness-first verification gates.
+
+## [10.0.0-rc.2] - 2026-09-22
+
+### Added
+- No-progress watchdog for fresh executor sessions with active-tool grace so legitimate long-running tools are not killed merely for being quiet.
+- Duplicate exploration guard and loop detector for repeated read/grep/glob-style calls without new workspace/evidence progress.
+- Runtime tool-output budgets for large grep/glob/read/repo-graph style outputs.
+- Durable pre-compaction checkpoints containing current task, runId, plan hash, workspace fingerprint, evidence pointers and a structured next action.
+- Post-compaction resume enforcement with automatic fresh-session recovery when the next action is not executed.
+- Provider failure classification and recovery for no-token, timeout, rate-limit, upstream, quota, auth and context-overflow failures.
+- Periodic lease supervisor that recovers expired running executors to an explicit retryable state.
+
+### Changed
+- Provider stalls retry once in a fresh session with the same model; repeated retryable provider failures use the configured escalation model/provider when available.
+- Stale lease recovery now records `retryable` instead of overloading terminal/logical `failed`.
+- Explicit `long/high-risk`, `high-risk`, and structured long/high-risk facts hard-override FAST/light routing to DEEP/heavy.
+- Package version is 10.0.0-rc.2.
+- npm tag releases publish prereleases under `next`; stable versions continue to use `latest`.
+
+### Fixed
+- A post-compaction session that only emits narrative text without executing the persisted next action is treated as stalled instead of silently completing.
+
+## [10.0.0-rc.1] - 2026-09-22
+
+### Added
+- Initial-input token telemetry and aggregate reporting for baseline-vs-UES evaluation.
+- Attempt-aware recovery policy with bounded context/skill escalation from initial execution to diagnosis and deep recovery.
+- Policy-aware FAST runtime routing that prioritizes direct domain/debug/review skills over generic orchestration.
+- Benchmark efficiency gates for initial input tokens and total tokens.
+- Reference-vs-candidate ablation CLI that requires pass-rate preservation and measurable initial-context reduction.
+
+### Changed
+- FAST context budget is 8k and STANDARD is 20k; DEEP remains 48k to preserve high-risk/long-horizon capability.
+- Always-loaded global engineering instructions are compressed while retaining exact-contract, evidence, verification, safety and durable-work invariants.
+- Retry context expands only after failure; repeated failures add graph/critic evidence instead of repeating speculative patches.
+- Adaptive model policy exposes recovery stage and does not silently downshift the configured/default executor tier for FAST tasks.
+- OpenCode V2 router metadata is versioned for the V10 policy-aware path.
+- Package version is 10.0.0-rc.1; npm publication/tagging is intentionally deferred until RC verification passes.
+
 ## [9.0.0] - 2026-09-22
 
 ### Added