dsh-super-code 0.3.10 → 0.4.0

package/docs/en/Acknowledgements.md

@@ -31,23 +31,17 @@ Design reference only, not a dependency. The UI is built on Harness's public API
 
 The bundled frontend Skill follows this format: `SKILL.md`, YAML metadata, and a `references/` directory. The model sees only the name and summary until the Skill is needed, then loads the body and relevant references. It coexists with project and user Skills. Parsing and loading use Harness's native support.
 
-Implementation: [super-code-frontend](../../presets/super-code/skills/super-code-frontend/SKILL.md)
+Implementation: [oil-ui](../../builtin/skills/oil-ui/SKILL.md)
 
-## Apple Human Interface Guidelines
+## oil-ui
 
-[Layout](https://developer.apple.com/design/human-interface-guidelines/layout) · [Accessibility](https://developer.apple.com/design/human-interface-guidelines/accessibility) · [Buttons](https://developer.apple.com/design/human-interface-guidelines/buttons) · [Feedback](https://developer.apple.com/design/human-interface-guidelines/feedback) · [Terms](https://developer.apple.com/support/terms/)
+[Source](https://github.com/oil-oil/oil-ui) · [MIT](../../builtin/skills/oil-ui/LICENSE) · [Pinned version and adaptations](../../builtin/skills/oil-ui/UPSTREAM.md)
 
-The frontend Skill adopts these principles: alignment expresses relationships, layouts adapt to containers and type size, control size and spacing reduce effort, interactions are predictable, feedback is close and proportionate. They're organized into three references: visual design, interaction states, and browser validation.
+Thanks to oil-oil for this open-source frontend design Skill. The plugin bundles oil-ui 0.32.0's workflow, references, comparison page, and helper scripts, loaded on demand through Harness's native Skill mechanism in place of the previous custom frontend Skill.
 
-Only principles are borrowed; each product keeps its own visual style. Native platform units aren't treated as CSS pixels, and materials, fonts, and icons aren't used as a template.
+The bundled copy is pinned to upstream commit `0d64a00f3a5c45bf7a30d35bded9b7c0fdff2128`. Its only adaptation replaces the standalone update entrypoint with plugin-managed updates; neither the upstream installer nor a paid plugin is included. The MIT license and original copyright notice are preserved in the package.
 
-## W3C WCAG 2.2
-
-[WCAG 2.2](https://www.w3.org/TR/WCAG22/) · [2.5.8 Target Size (Minimum)](https://www.w3.org/WAI/WCAG22/Understanding/target-size-minimum.html)
-
-When checking Web controls, the frontend Skill uses 2.5.8's 24 × 24 CSS pixel criterion and its spacing exceptions, alongside real click, keyboard, and content checks. Passing the numbers does not make the whole interface accessible.
-
-Implementation: [interaction states](../../presets/super-code/skills/super-code-frontend/references/interaction-states.md), [browser validation](../../presets/super-code/skills/super-code-frontend/references/browser-validation.md)
+The comparison page also bundles [Instrument Serif](https://github.com/Instrument/instrument-serif), under [SIL OFL 1.1](../../builtin/skills/oil-ui/assets/fonts/instrument-serif/OFL.txt), for comparison-page typography.
 
 ## SWE-bench
 
@@ -74,3 +68,21 @@ Full list in [package.json](../../package.json); locked versions in [package-loc
 ## License
 
 This project uses [Apache-2.0](../../LICENSE). The projects, specifications, design resources, and datasets above keep their own copyrights and licenses as stated upstream (LICENSE, NOTICE, and dataset docs). A design reference does not imply a dependency, endorsement, or feature equivalence.
+
+## OpenSpec
+
+[Source](https://github.com/Fission-AI/OpenSpec) · [MIT](../../builtin/skills/openspec/LICENSE) · [Upstream](../../builtin/skills/openspec/UPSTREAM.md)
+
+Thanks to the OpenSpec contributors for the complete specification-driven workflow Skill distribution. The plugin preserves the original prompts and licenses and loads them on demand through DSH.
+
+## Matt Pocock Skills
+
+[Source](https://github.com/mattpocock/skills) · [MIT](../../builtin/skills/mattpocock/LICENSE) · [Upstream](../../builtin/skills/mattpocock/UPSTREAM.md)
+
+Thanks to Matt Pocock and contributors for the complete engineering and productivity Skill distribution. The plugin includes the official plugin inventory with its references, templates and invocation rules; it does not run setup on activation.
+
+## PStack
+
+[Source](https://github.com/cursor/plugins/tree/main/pstack) · [MIT](../../builtin/skills/pstack/LICENSE) · [Upstream](../../builtin/skills/pstack/UPSTREAM.md)
+
+Thanks to Lauren Tan and the Cursor plugin contributors for PStack. The plugin preserves its 58 standard Skills, invocation rules, references and scripts, loaded on demand through native DSH Skills and slash commands. It does not register Cursor automations or run setup on activation. Some workflows still require Cursor capabilities or external tools; see the upstream notes.

package/docs/en/Changelog.md

@@ -4,6 +4,76 @@
 
 Versions are `major.performance.maintenance`. The second number changes when coding results or efficiency improve; the third covers fixes, UI, and compatibility. See [Compatibility](Compatibility.md) for what each release was tested on.
 
+## 0.4.0 · 2026-11-12
+
+**Task and memory tools**
+
+- Simplified calls to `super_code_task` and `super_code_memory`, with action-specific field guidance, examples, and recovery instructions to reduce trial and error.
+- Read a named project-memory topic without creating or switching the focused task. Writes retain source checks and revision-based conflict protection; historical information does not replace checking current code.
+
+**Preset Skills and workflows**
+
+- Added standalone grill-me alongside OpenSpec, oil-ui, Matt Pocock Skills, and PStack, grouped into Skills and workflows. All five default off, and saved choices are preserved.
+- Enabled packages sync their complete resources to the DSH Skill directory and take effect in existing conversations. Each entry shows its version and a short usage hint, with a shortcut to open the directory. Existing user files and edits are protected; conflicts are reported and can be retried.
+- Uses the native `/` menu and real Skill names. Search by package names such as `/pstack` or `/mattpocock`, then select a specific command. Disabled packages no longer supply commands or bodies; content already loaded into context is not withdrawn.
+- OpenSpec's command environment ships with the plugin. Toggling a package does not initialize a project, install project dependencies, or modify project files; initialization runs only when requested.
+- PStack retains upstream workflows and resources, with adaptations for DSH tools, child Agents, approvals, and image handling. Workflow adaptation targets Harness `0.2.1-alpha.1` and later.
+- Bundled and adapted sources are published under `builtin/skills`, with upstream versions, provenance, and licenses.
+
+**UI, compatibility, and fixes**
+
+- Improved Settings autosave, draft retention across tabs, reconnection, concurrent edits, and recovery after failed saves.
+- Improved task, execution, tool-status, and changed-file displays while retaining circular Agent nodes and the existing memory page.
+- Fixed cancellation, reconnection, and repeated enable/disable issues in preset installation and uninstall recovery, MCP connections and approvals, deletion protection, and the Windows command environment.
+- Updated compatibility records for newer Harness versions. See [Compatibility](Compatibility.md) for version and platform coverage, distinguishing SDK checks from full-host acceptance; interface tests do not establish that every workflow and platform has passed.
+
+- Added version, license, and total npm download badges, with Star and npm trend charts refreshed daily.
+
+See the [README](../../README.en.md) and [evaluation notes](EvaluationResults.md) for results and statistical limits. Skill installation, UI, and compatibility tests are not evidence of coding performance.
+
+## 0.3.11 · 2026-11-06
+
+**Uninstall recovery and preset management**
+
+- Fixed Super Code conversations failing to continue after uninstall: registered conversations with matching ownership continue with the official Standard preset and show one switch notice. Existing parent/child history, model, directory, and permissions are preserved.
+- Fixed stale deletion-protection rules in official Minimal after protection is disabled, official-preset coverage is removed, or the plugin is disabled or uninstalled. Changing levels no longer stacks duplicate rules. Existing summaries remain intact, followed by an explicit revocation notice.
+- Fixed stale preset recovery during asynchronous reads, user selection changes, plugin re-enablement, and module replacement. Retired instances no longer overwrite newer records or take over the service again.
+- Fixed installation/reinstallation rollback and locale synchronization overwriting external changes or losing ownership records. User edits and symbolic links are preserved, invalid records are rejected, and conflicting writes stop.
+
+**Task sidebar, execution records, and usage**
+
+- Fixed the task sidebar showing only the current request. It retains the latest 20 earlier requests and their recorded steps, with a notice when display limits are exceeded.
+- Fixed unrendered execution details by using the host's Markdown and code components, preserving literal user/tool output and allowing long content to expand.
+- Completed PTC child-tool call, result, failure, and terminal-error records, with correct interrupted status after crash recovery. Successful file-change evidence survives slow calls leaving the recent activity window.
+- Fixed duplicate history reads, shifting boundaries, and lost reading positions during pagination and reconnection. Failed history reads can be retried, and closed views ignore late responses.
+- Fixed inherited fork history being counted again in child and tree usage, and distinct model IDs being merged when they contain separator characters. Old caches are rebuilt.
+- Fixed all sidebar sections collapsing: exactly one of Tasks, Collaborative execution, and Memory stays open, with Tasks as the default. Switching preserves reading positions and drafts.
+
+**Settings, budgets, and memory**
+
+- Fixed Home reverting to stale settings when a save succeeds but the following refresh fails. Reset now waits with controls locked during pending autosave, preventing duplicate submissions and revision conflicts.
+- Fixed cancellation during a model-information lookup being ignored by automatic compaction or still allowing tool-record pruning. Existing compaction policy and budget defaults are preserved.
+- Fixed retained legacy-memory drafts using a retired API after reconnection. Drafts are kept, explicit retries use the current API, and in-flight writes are not replayed automatically.
+- Fixed empty topics continuing to consume the 200-topic allowance after their last entry is deleted. Capacity counts non-empty topics while retaining concurrent-edit checks, without adding content version history.
+
+**MCP connection**
+
+- Fixed asynchronous preparation using revoked directory access or disabled settings. Current authorization and cancellation are checked again before conversation creation, model selection, and prompt submission.
+- Fixed `newConversation: false` creating a conversation without `taskId`. The caller must now choose a specific conversation first.
+- Fixed failed create, continuation, or cancellation saves changing in-memory state prematurely, consuming capacity, or causing delayed cancellation. Preparation retains previous-request tracking; restart and reconnection never automatically replay saved requests.
+- Fixed late approval being accepted after an approval wait is cancelled. One-time decisions recheck cancellation, directory access, and settings; undecided requests can return to DSH.
+- Fixed directory requests being approved after path replacement, expiry during validation, or withdrawal. Reusing an old request cannot restore revoked access.
+- Fixed symbolic-link aliases producing duplicate allowed directories that reappear after removal. Each real directory appears once, and unavailable entries remain removable.
+- Added missing public PTC child-call records to `dsh_read`. Child-tool activity now counts toward progress observation instead of being mistaken for a long idle period.
+- Fixed connection-capacity, port, and lock leaks during initialization, token rotation, shutdown, and unload races. Successful saves return the actual settings; cleanup failures are explicit and retryable without resending tasks.
+
+**Commands and runtime**
+
+- Fixed missed and false deletion detections in some shell wrappers, `find`, `xargs`, and PowerShell launch arguments. Existing protection levels and host approval still decide execution.
+- Fixed Windows shell filters, official-budget adapters, and PTC wrappers surviving unload or overwriting other wrappers. Plugin resources are released after asynchronous initialization settles, preserving host and other-plugin state.
+
+Fixes have regression checks scoped to their changes. Version contracts, frontend checks, and full-host acceptance are recorded separately; this does not mean every version on every platform has passed full-process tests. No new 100-task effectiveness evaluation was run.
+
 ## 0.3.10 · 2026-11-03
 
 **Windows command environment**

package/docs/en/Compatibility.md

@@ -2,9 +2,13 @@
 
 [中文](../Compatibility.md) · English
 
-Web is verified from Harness `0.1.5-alpha.1` through `0.2.1-rc.4`, with incremental rc.4 coverage detailed below. Official Desktop is verified through `0.2.1-rc.4` (x64 archive under Rosetta), and NEXT at `2.0.19-next` (bundled `0.2.1-rc.2`).
+Supports Harness `0.1.5-alpha.1` through `0.2.2-rc.2`, including `0.2.2-alpha.3`. Version 0.4.0 has been checked against all 27 complete Web hosts in this range, official macOS arm64 Desktop `0.2.2-rc.2`, and NEXT `2.0.22-next` (bundled `0.2.2-rc.1`). This round's scope and limits are recorded below.
 
-The bundled Windows Bash in 0.3.10 also passes full-process Windows acceptance on official Desktop `0.2.1-rc.5` and NEXT `2.0.21-next` (bundled `0.2.1-rc.4`), within the scope below.
+Full-process Windows coverage currently reaches official Desktop `0.2.1-rc.5` and NEXT `2.0.21-next` (bundled `0.2.1-rc.4`). New macOS/Web results do not extend that Windows scope.
+
+The records below distinguish each code baseline and test layer. Component checks, static-frontend fixtures, and earlier development-package results do not establish a complete rerun of every version and platform.
+
+The new PStack workflow adaptation starts at `0.2.1-alpha.1`; earlier hosts are outside its scope. Older-version records below describe historical coverage of other plugin features.
 
 ## Verified hosts
 
@@ -53,6 +57,78 @@ Under npm rules, a range like `>=0.1.5-alpha.1 <0.3.0` excludes prereleases such
 
 `zod`, `js-yaml`, and `@modelcontextprotocol/sdk` are runtime dependencies and are not bundled. The Windows shell also packages separately licensed programs that run as subprocesses; see [runtime notes](../../runtime/windows/README.md) for sources, licenses, and corresponding source.
 
+## November 12, 2026 · 0.4.0 compatibility rerun
+
+The existing peer dependency ranges and minimum `0.1.5-alpha.1` remain unchanged. Every version uses the same 0.4.0 candidate; no plugin runtime changes were needed for these checks.
+
+| Layer | Scope | Result |
+|---|---|---|
+| Complete Web hosts | All 27 exact published versions, `0.1.5-alpha.1` through `0.2.2-rc.2` | 12 groups per version; 324 pass |
+| Actual SDK and disk storage | The same 27 versions | Six groups per version; 162 pass |
+| Native Skill loading | The same 27 versions | 24 checks and 98 body loads per version pass; duplicate names follow provider precedence |
+| Native menu contract | The same 27 versions | Pass; `0.1.5` lacks package-query expansion, but individual Skill names remain selectable |
+| Browser interactions | Official React, controls and themes from all 27 versions, with fixture Remote services | 15 groups per version; 405 pass without page errors |
+| Official macOS arm64 Desktop | `0.2.2-rc.2` | 13 complete-host groups and four actual UI checks pass |
+| DSH NEXT · macOS arm64 | `2.0.22-next`, bundled `0.2.2-rc.1` | 13 complete-host groups and four actual UI checks pass |
+
+- **Complete-host checks**: install the candidate, verify preset versions, independent budget/protection saves, five default-off Skill packages enabling and disabling in existing conversations, default MCP settings, manual memory CRUD and concurrency conflicts, native/PTC task and memory calls, complete process restarts, original-conversation continuation and memory recovery.
+- **UI checks**: six settings tabs, Skill versions and switches, failed saves, budget validation, connection settings, Chinese/English, light/dark themes, narrow layouts, 10,000-row task virtualization, execution-tree animation and legacy memory maintenance. Both actual desktops additionally verify settings loading, matching versions, five default-off packages, switch autosave and disabled Windows command-environment controls on macOS.
+- **Provenance and baseline**: the official archive matches the feed SHA-512 and passes deep signature verification; all 333 DSH components are rc.2. NEXT has a valid signature and all 330 DSH components are rc.1. All 1154 unit tests, type checking and build/pack pass. Acceptance scripts were corrected for concurrent completion order, upstream theme packaging and actual page structure; original failed records remain preserved.
+- **Limits**: complete hosts use a local deterministic model adapter with zero external-model calls. This does not establish model effectiveness, acceptance of all 58 PStack workflows or every feature branch. This round does not repeat Windows, Linux, every historical native desktop, native directory pickers or physical-uninstall recovery; those retain their historical evidence identities. Full PStack workflow adaptation still starts at `0.2.1-alpha.1`.
+
+## November 11, 2026 · PStack DSH adaptation
+
+- Batch six removes adapter-added command bans, mandatory API checks and repeated snapshot audits. Four role templates now match upstream exactly, and delegation translation preserves the upstream read-only requirement. Read-only roles may use commands without side effects, but may not write files or change external state. Command capture also fixes sparse-argv validation and post-submission argument mutation. All 925 regression tests and 143 resource checks across 13 versions pass; real minimum/latest Web capture scenarios each pass nine independent checks. Live workflow children still redirect output into temporary files, so read-only scope adherence remains unresolved.
+- The scope contains all 13 npm releases from `0.2.1-alpha.1` through the current latest `0.2.2-rc.1`, including `0.2.2-alpha.3`. Native catalogs, all 58 bodies, slash injection, invocation permissions and same-name precedence pass, with batch five adding command-capture resources for 143 checks; task enum diagnostics pass another 39 checks. Native subagent contracts pass 65 checks; live switches in existing browser conversations pass 169. Browser checks use real frontend modules with fixtures at service boundaries, not 13 complete desktop processes.
+- On November 11, 2026, all 58 entrypoints ran through native slash injection and real model calls in isolated macOS arm64 Web `0.2.2-rc.1`, using `tokenhub/deepseek/Evaluation model A`. Each entrypoint had an independent scenario. Twelve scenarios were rerun, and two received clarifications; repeats are not added to the 58-entry denominator.
+- First-round acceptance: 36 pass, 17 pass with qualifications, four fail and one partial. Read-only children in `how`, `teach` and `create-formal-verification-skill` still wrote global temporary files. `principle-outcome-oriented-execution` retained `add` twice and passed after the target was clarified as `sum({left,right})`; ambiguity between the original wording and the acceptance contract needs separate review. `ketchup-video` encoded and decoded a real video, but the test route omitted the image-input declaration and the workflow did not complete its visual review. Execution completion is not acceptance.
+- Image diagnosis corrected: the model does support images. With `input: ["text", "image"]` declared in the isolated route, direct image attachments and native `read_image` on both `0.2.1-alpha.1` and `0.2.2-rc.1` let the live model read a random code and count shapes present only in the pixels. Provider requests contained image content. A text-only configuration rejected images before sending. The Chinese rc.1 Web UI also passed native selection, preview, removal, paste, drop and history display; this round did not exercise native desktop image interactions across platforms.
+- All 22 entries without full acceptance were rerun. The video workflow completed 14 actual image reads, including decoded frames from every slide; eight CLI samples and four decoded images were independently checked. The complete `statechart` workflow also passed real TLC and code replay. Independent baseline checks passed 56/57; the remaining check compares an ambiguously worded API task against a fixed contract and needs separate interpretation. Five additional checks passed. Read-only children still wrote global temporary files, and cleanup used `rm` and process-name matching to stop encoders. Image transport and baseline checks do not erase these defects, so all 58 workflows cannot be reported as fully accepted.
+- Three read-only/audit workflows were tested again, passing seven baseline checks. An independent copy of the formal-verification artifact passed three additional normal, invalid-argument and mutation/restoration checks. The audit reviewer made no writes this time, and its independent review passed with the snapshot coverage limits disclosed. The `teach` investigator and synthesizer still wrote global temporary files. Prompt rules therefore do not provide a hard read-only boundary, and these issues are not all resolved.
+- This round corrected three global temporary-path examples and child-scope propagation guidance. It also fixed the installed `statechart` helper: Node refuses TypeScript stripping under `node_modules`, so packaging now emits JavaScript while retaining upstream sources. The actual installed wrapper passed TLC, a concrete authorization counterexample and code replay under a Unicode path with spaces. The MCP bot UI exercised real new/reused conversations, stable request IDs, restart, rejection branches, paged history and session termination; live approval waiting and remote deployment were not exercised.
+- The distribution retains 58 entrypoints, 23 playbooks and 558 originals, with 592 generated files, including audit snapshots, stable submission admission and project-local command capture. Original hashes, generated bytes and helper permissions are checked. All 919 batch-five tests, type checking and build pass; the earlier 909-test batch and 19 focused follow-up checks remain recorded. The initial 58-scenario independent checks total 155: 154 pass and the API migration assertion fails. These checks do not replace phase and scope review.
+- A third batch reruns eight priority entrypoints with 20 passing baseline assertions. The converter now preserves why's code-anchor step, and source-only contracts are inside the actual role templates. Three video slides and two transitions were independently decoded and viewed; 16 checks pass, without permanent shell cleanup or shared-name process termination. Eight snapshot-helper checks cover append/tamper/CRLF/Unicode behavior; the complete 23-row trail receives 12 provenance/behavior checks plus per-row semantic review. Structural validity is not claim verification.
+- Live MCP approval and cancellation now pass six checks; the browser observes `waiting_confirmation → cancelling → cancelled`, and the pending probe never executes. Acceptance found a changed-content cache bug in the generated example backend. After a manual repair, 13 checks pass; the product workflow now specifies fingerprints, serialized admission and original-input retries. This is neither a newly found core-service defect nor an autonomous first-pass success. Remote deployment and live injection of every error status remain untested.
+- Minimum `0.2.1-alpha.1` reruns a seven-file how scenario with the final runtime candidate: 14 independent checks pass, native command results are used directly, the source/tree stays unchanged, and two children use only read/glob/grep/send_message. Model adherence limits remain: teach/show use read-only cat/ls, and the minimum run repeats an explainer and attempts a capture write that fails with ENOENT; its handoff also undercounts children. The final package then reruns managed and synchronous return branches on the minimum host: 15 checks pass, with one delegate per case, no capture/write attempt or redispatch, and same-call final-text extraction for the synchronous result. Earlier duplicate/undercount evidence remains; the synchronous reply still exceeds its requested length. Neither hard isolation nor universal prompt compliance is established.
+- Batch four adds a dependency-free stable submission helper to make-bot-ui, covering concurrent IDs, content conflicts, restarts, persistence failures and independent return values. The generated example completes 92 self-checks. Acceptance then finds interrupted incorrectly treated as settled and failed transports closed without DELETE. After diagnostic continuation in the original model session, 91 deterministic fixture checks, 13 independent UI checks and two recovery checks pass. The session then reaches the acceptance harness's cumulative 100-request limit without its final handoff, so this is not an autonomous full-pass claim. The final packaged helper separately passes 18 independent checks with real MCP/model calls: new and explicitly reused conversations, per-request prior/current results, refusals and session release. Fixture states are not real network-failure injection.
+- The fourth audit covers all work/correction rows 1–28; final rows 29–30 contain review metadata only. Twenty-five snapshot, actual CLI and provenance checks pass. Replaced evidence, incorrect counts and causal descriptions remain disclosed; the reviewer verdict is PASS WITH FLAGS. Teach's four children make no writes but one still uses ls under a narrower no-shell contract. Bot development writes four global temporary responses and issues a cleanup command containing an rm fallback whose actual branch is unknown. A narrow live scenario after project-local evidence guidance gets all three HTTP refusals right but still writes a global status-probe file; its scope check remains failed. Functional checks do not establish universal scope adherence.
+- Batch five retains unique project-local capture directories: all 34 audit captures re-verify and 14 independent provenance/CLI checks pass. The original session appends corrections for mistyped hashes and row counts; its reviewer still executes a read-only shell command, so the existing no-shell contract remains failed. The repaired MCP example checkpoint completes 96 real integration checks, 91 explicitly simulated fault checks and its final handoff, without global temporary responses or overwritten command evidence. Nineteen of 20 independent checks pass; the sole failure is that role contract. A separate real minimum-host how/capture scenario passes 16 checks for Unicode paths with spaces, literal argv, exit 0/7, exact streams, byte verification, unchanged source and a read/search-only child. These targeted checks do not establish full acceptance of all 58 workflows.
+- The 58-scenario run covers only that macOS Web host and model, not Windows/Linux, every desktop release, all 23 playbook branches, cross-model evaluation, GitHub publication or the optional Bun watcher. Earlier minimum-host configuration/delegation evidence is retained separately; the new how run does not cover every minimum-host workflow. No effectiveness or performance improvement is established.
+
+- The completeness review restores setup’s eight stages and 17 role categories, independent reflect synthesis, incremental history, simulator/cache branches, and cause-based retries/late-result handling. Reasoning budget, candidate count and concurrency remain distinct. Missing public model discovery means inheritance, not scanning private configuration. Long runs assume neither a scheduler nor a private agent store.
+
+- Follow-up live-model checks: latest-host setup preserves the full role map and public capability-discovery boundary; reflect runs three independent reviews plus a fourth synthesizer, routes structural proposals to Backlog and makes no Skill edits before approval. Minimum `0.2.1-alpha.1` rechecks the new setup’s 17 roles, separate budget/panel/concurrency values and existing notes. Initial discovery/classification failures and acceptance-script format assumptions remain recorded; a completion marker alone is not acceptance evidence.
+
+## November 9, 2026 · Latest-host rerun
+
+This run uses `main@8d28e47`, including the oil-ui `0.32.0` replacement. The fixed test package has SHA-256 `3b19a6c04b960202d87e13bb17e59085f7378503aaf8316a9c0d78220f6c50d5`. It is not the original v0.3.11 Release archive; no version bump or republication was performed.
+
+At checking time, official [npm](https://www.npmjs.com/package/@deepseek-ai/dsh) `latest`/`next` and the [most recent GitHub Release](https://github.com/deepseek-ai/deepseek-harness/releases/tag/dsh-v0.2.2-rc.1) are all `0.2.2-rc.1`, released November 4, 2026. The alpha tag is `0.2.2-alpha.3`; no later release was found. rc.1 changes Loader, tool execution, and session projections but retains the legacy projection registration used by the plugin. No plugin runtime change is required within the tested scope below.
+
+| Complete host | Actual version | Result |
+|---|---|---|
+| Official Web | `0.2.2-rc.1` | Pass |
+| Official Web · alpha channel | `0.2.2-alpha.3` | Pass |
+| Official Desktop · macOS arm64 | `0.2.2-rc.1` | Pass |
+| DSH NEXT · macOS universal archive | `2.0.22-next`, bundled `0.2.2-rc.1` | Pass |
+
+- **Complete flow**: the same package runs through actual Loader, Agent, tools, PTC, and session storage. Checks cover standard/PTC/minimal/Super Code/renamed-copy presets, installation and reinstall backups, budgets and deletion settings, parent/child histories, compaction, legacy memory CRUD/conflicts/scoping, cold restarts, plugin disable/re-enable, and physical removal followed by preserved history and official-standard continuation. Super Code shows one recovery notice; minimal excludes revoked plugin rules. The real deferred Skill tool loads oil-ui, verifying its body, provider, relative resources, and installed version.
+- **MCP boundaries**: each passing full host additionally checks port conflicts, token rotation, and directory consent, plus 39 checks across admission, persistence failures, approval lifecycle, directory changes, and storage locks. Coverage includes cancellation, disconnect, revocation, idempotency, and no replay after restart.
+- **Frontend and baseline**: official React, controls, and themes from `0.1.5-alpha.1`, `0.2.2-rc.1`, and `0.2.2-alpha.3` each pass 14 interaction groups with fixture services. These cover five settings tabs, autosave/failures, custom dropdowns, Chinese/English, themes, 440px settings layouts, 10,000-row virtualization, circle animation, exactly one open section, and legacy memory CRUD, without page errors. These are not three complete hosts. All 560 unit tests, type checking, and build pass.
+- **Provenance and versions**: the official arm64 ZIP matches its nightly feed SHA-512 and the original application passes deep signature verification; all 321 DSH components are rc.1. The NEXT download channel points to `2.0.22-next`, all 330 DSH components are rc.1, and the original application passes deep signature verification. Its API supplies no public digest; local SHA-256 only pins the tested sample. Both Web graphs contain 320 DSH components, all at their target version.
+- **Limits**: full hosts use a local deterministic model adapter, without external-model calls or effectiveness evaluation. This run does not repeat every historical complete host, latest native Windows/Linux processes, or native directory picking. The rc.1 host requires Node `>=24.9` when upgrading Web; the plugin retains its Node range for older hosts.
+
+## 0.3.11 · 2026-11-06
+
+This release includes successive fixes from November 4–6. Results below apply to the code tested at each stage; full-host results from an earlier development package are not presented as a complete platform rerun of the final release package.
+
+- **Release candidate**: all 560 tests, type checking, build, and client syntax checks pass. The package, lockfile, directory preset, and declarative preset all use 0.3.11.
+- **Version contracts and frontends**: 26 exact published Harness versions, from `0.1.5-alpha.1` through `0.2.2-alpha.3`, cover the listed 0.1.5, 0.1.6, 0.1.7, 0.2.0, and 0.2.1 alpha/rc releases, plus `0.2.2-alpha.2`, `0.2.2-rc.1`, and `0.2.2-alpha.3`. Change-specific checks cover Session/LLM recovery, task projections, usage, compaction cancellation, and legacy memory. Frontend checks use actual official React and plugin components with fixture Remote boundaries; they are not 26 complete hosts.
+- **Recent targeted evidence**: PTC history has 26-version contract and frontend checks; fork usage has 157 contract checks across 26 versions; compaction cancellation has 170 checks; directory aliases have 156 frontend checks; memory capacity has 156 storage-contract checks plus frontend interactions on the earliest/latest versions; settings reset has 182 frontend checks across 26 versions. All pass. Counts from separate batches are not added up as distinct hosts.
+- **Full Web and macOS hosts**: uninstall-recovery and MCP lifecycle fixes were checked on Web `0.1.5-alpha.1`, `0.2.1-rc.5`, `0.2.2-rc.1`, and `0.2.2-alpha.3`, official macOS Desktop `0.2.1-rc.5`, and NEXT `2.0.19-next` (bundled `0.2.1-rc.2`). Later task-history/rendering fixes additionally checked real tools, three turns, and cold restart on the earliest/latest Web hosts and official macOS rc.5. Subsequent usage, compaction, directory, memory-capacity, and reset fixes did not rerun the full desktop matrix.
+- **Windows and Linux**: an earlier stage ran actual official x64 Desktop rc.5 and NEXT `2.0.21-next` (bundled rc.4) in a Windows 11 ARM64 VM, with 37 passing groups each for MCP, Bash, PowerShell, deletion protection, physical uninstall, cold restart, and history continuation. All 118 Windows regression tests pass, with three POSIX-specific skips, plus 22 PowerShell 5/7 comparisons. A staged non-root Linux ARM64 full suite passes 527 tests. Later fixes did not rerun those complete operating-system suites.
+- **Preservation and limits**: within the accepted scenarios, parent/child history remains readable; Super Code continues with official Standard after uninstall and shows one notice, while Minimal revokes stale plugin rules. Public memory remains legacy CRUD. Host models use deterministic adapters, with no external-model calls, new 100-task evaluation, Safari-specific acceptance, or per-version historical native Windows desktop acceptance.
+
 ## 0.3.10 · 2026-11-03
 
 These Windows Bash checks ran at successive stages on November 2–3, 2026. Each batch retains its code baseline and scope; historical release records below are unchanged.

package/docs/en/EvaluationResults.md

@@ -1,10 +1,12 @@
 # Evaluation Results
 
+Model names are anonymized; lettered labels distinguish evaluation models.
+
 [中文](../EvaluationResults.md) · English
 
 ## Pro100: continuous-environment comparison
 
-All runs use Harness `0.2.1-rc.2`, `deepseek-v4.1-flash-1017`, high reasoning effort, and legacy budgets. Each group reuses its environment continuously across tasks. These are full 100-task results, sorted by tasks passed.
+All runs use Harness `0.2.1-rc.2`, `Evaluation model B`, high reasoning effort, and legacy budgets. Each group reuses its environment continuously across tasks. These are full 100-task results, sorted by tasks passed.
 
 | Configuration | Passed/valid | Mean total tokens (M) | Mean time (min) | Mean model calls | Mean tool calls | Cache rate |
 |---|---:|---:|---:|---:|---:|---:|
@@ -38,8 +40,8 @@ Comparisons on the same host keep model, reasoning effort, and budgets fixed whi
 
 ### Historical process
 
-1. **Old-host baseline**: official presets and plugin versions were evaluated on DSH `0.1.5-rc.2` with `deepseek-v4.1-flash`. The Super Code `0.3.0` candidate passed 56 tasks, above the best official result (Standard + high, 51).
-2. **Regression after the host upgrade**: on DSH `0.2.1-rc.2` with `deepseek-v4.1-flash-1017`, under default budgets Standard fell to 46 and Super Code `0.3.4` to 41, below Standard.
+1. **Old-host baseline**: official presets and plugin versions were evaluated on DSH `0.1.5-rc.2` with `Evaluation model B`. The Super Code `0.3.0` candidate passed 56 tasks, above the best official result (Standard + high, 51).
+2. **Regression after the host upgrade**: on DSH `0.2.1-rc.2` with `Evaluation model B`, under default budgets Standard fell to 46 and Super Code `0.3.4` to 41, below Standard.
 3. **Root-cause analysis**: a host probe showed that the new host tightened context budgets (see [Contents of the legacy budgets](#contents-of-the-legacy-budgets)), identifying budgets as one cause of the regression.
 4. **Retest with legacy budgets**: both sides switched to legacy budgets. Standard stayed at 46; Super Code `0.3.5` rose to 49. Based on this, the plugin uses legacy budgets by default from `0.3.6`.
 5. **Cross-model check**: compared with native ZCode on `glm-5.3-flash` to test whether the approach depends on a specific model.
@@ -56,7 +58,7 @@ Comparisons on the same host keep model, reasoning effort, and budgets fixed whi
 
 ## New DSH and budgets
 
-DSH `0.2.1-rc.2` · model `deepseek-v4.1-flash-1017` (API ID `deepseek/deepseek-flash`) · high.
+DSH `0.2.1-rc.2` · model `Evaluation model B` (API ID `deepseek/Evaluation model A`) · high.
 
 | Metric (per-task mean) | Standard · default | Super Code 0.3.4 · default | Standard · legacy | Super Code 0.3.5 · legacy |
 |---|---:|---:|---:|---:|
@@ -99,7 +101,7 @@ The same task has passed in one run and failed in another. Because configuration
 
 ## Official presets
 
-DSH `0.1.5-rc.2` · `deepseek-v4.1-flash`.
+DSH `0.1.5-rc.2` · `Evaluation model B`.
 
 | Metric (per-task mean) | Minimal + high | Minimal + max | Standard + high | Standard + max | PTC + high | Create + high |
 |---|---:|---:|---:|---:|---:|---:|
@@ -126,7 +128,7 @@ DSH `0.1.5-rc.2` · `deepseek-v4.1-flash`.
 
 ## Historical versions
 
-DSH `0.1.5-rc.2` · `deepseek-v4.1-flash` · high. Times use the historical raw convention (see [Methodology](#methodology)); don't compare them with the net times above.
+DSH `0.1.5-rc.2` · `Evaluation model B` · high. Times use the historical raw convention (see [Methodology](#methodology)); don't compare them with the net times above.
 
 | Metric (per-task mean) | Minimal + high | 0.0.9 | 0.1.2 | 0.2.0 layered memory | 0.3.0 Batch.3 |
 |---|---:|---:|---:|---:|---:|

package/docs/en/UsageGuide.md

@@ -22,6 +22,12 @@ The plugin has no model settings of its own. If you already have a working model
 
 Child Agents use the parent's model by default. Harness manages permissions and tools.
 
+### Image input
+
+For a model that supports vision, enable **Image** under **Input types** in Harness **Settings → Models**, then select that model in the conversation. Add an image through the composer's **+** menu, paste a screenshot, or drag an image in. Preview or remove it before sending. Super Code uses the host's attachment, resizing and sending pipeline.
+
+The model reads project images through native `read_image`, supporting PNG, JPEG, WebP and GIF. Screenshots and decoded video frames require actual image inspection; a path or text summary is not a visual check. If the host says the model does not declare image input, check the selected model's capability setting first. That error does not establish that the underlying model lacks vision. Keep image input disabled for text-only models.
+
 ## Can't find the preset
 
 First make sure the plugin is enabled, then check your host version against [Compatibility](Compatibility.md). Then:
@@ -74,13 +80,27 @@ Settings shows the **plugin version** and the **preset version** separately.
 
 **Safety**:
 
-- Only presets verifiably installed by this plugin are replaced (directory: install record, UUID, path, directory identity; declarative: source marker or install record). Unmarked old copies and same-name declarations are left alone.
+- Directory preset status, language synchronization, and reinstallation verify the install record, UUID, path, and directory identity. Declarative presets verify the source marker or install record. Older copies and same-name declarations with unverified ownership are left unchanged and remain readable.
 - The old content is backed up in `.super-code-backup-*`: under the user preset root for directory hosts, under the current profile for declarative hosts (ownership records in `.super-code-presets.json`).
-- A failed install restores automatically. After an unexpected crash, restore manually from the backup.
+- Failed installs roll back their writes and check directory identity before cleanup; substituted directories are preserved. The host then attempts to reload the restored configuration. An originally absent configuration file stays absent. After a crash, or if ownership records remain unwritable, inspect the backup before restoring manually.
+
+### Continuing after uninstall
+
+After loading a plugin version with recovery support, you can still read existing history when Super Code is uninstalled. Former Super Code conversations continue with the official Standard preset and display a one-time switch notice. The model, working directory and host permissions are preserved. Conversations already using an official preset keep that preset.
+
+Turning off deletion protection, disabling its official-preset scope, or disabling or uninstalling the plugin revokes previously injected deletion rules on the next continuation, including Minimal's rules. Switching between strict and relaxed modes retires the previous rule; cold restarts do not add the same rule again. Original history and your own deletion instructions are preserved. If an older rule has already been compacted into a summary, an explicit revocation is added without discarding the summary's other context.
+
+Disabling or uninstalling the plugin also releases its command filters, budget adapters, and MCP service. Host permission restrictions and changes owned by other plugins are preserved.
+
+If the plugin is reinstalled or the selected preset changes during recovery, the current installation and latest selection take precedence over the fallback. Recovery module updates retain existing conversation registrations and stop late writes from the retired module.
+
+If another process changes local recovery or preset ownership records while DSH is running, the plugin stops writing and preserves those files. Check the files, then restart DSH to reload them. An unrelated broken preset does not prevent other usable conversations from being registered.
+
+The profile retains `.super-code-recovery` and its configuration entry so recovery works after a cold start without the npm package. It restores registered conversations and revokes old plugin deletion rules; it contains no Super Code tools or model configuration. An upgrade registers older conversations whose preset definitions still exist. If a preset was deleted or renamed before registration, first restore its definition from the installation backup, then reinstall and load the plugin to register the conversation. If Standard is also unavailable, the error is preserved rather than silently choosing another preset.
 
 ## Sidebar
 
-In the right sidebar choose **Super Code**. Three sections, at most one open at a time, Tasks by default. Click the open heading to collapse all sections. Switching keeps reading position, filters, and drafts. The UI follows Harness's Chinese/English setting.
+In the right sidebar choose **Super Code**. Exactly one of the three sections stays open, Tasks by default. Clicking the open heading keeps it expanded. Switching keeps reading position, filters, and drafts. The UI follows Harness's Chinese/English setting.
 
 ### Tasks
 
@@ -88,6 +108,7 @@ Read-only. Shows what's already recorded in the session:
 
 - **Goal**: the recorded task goal, or the user's original message if there is none.
 - **Steps**: todos the model wrote down, shown as done/total. No steps means no invented plan or progress.
+- **Earlier requests**: prior user requests and their recorded steps stay visible as history, separate from current progress.
 - **Tool calls**: success, failure, and running counts grouped by tool.
 - **Files**: successful file changes that can be confirmed from tool arguments or diffs, including nested PTC calls.
 
@@ -100,11 +121,11 @@ Notes:
 
 ### Agent execution
 
-The Agent tree, with status, usage, and cache rate for each Agent.
+The Agent tree, with status, usage, and cache rate for each Agent. Usage counts only calls made by that Agent; forked history is not charged again to the child or the tree total.
 
 - Drag empty space to pan; zoom with the wheel or buttons.
 - Click a node to open its details in a new sidebar tab. The main conversation stays put.
-- Details are shown as a chat: user task, Agent replies, expandable tool calls and results. The header shows status, tokens, and cache; open **Details** for more usage, task records, and memory activity.
+- Details are shown as a chat: user task, Agent replies, expandable tool calls and results, including PTC sub-calls and terminal failures. Harness's Markdown and code components render lists, tables, and highlighted code, while tool arguments and output remain literal. The header shows status, tokens, and cache; open **Details** for more usage, task records, and memory activity.
 - Running animations respect the system's reduced-motion setting.
 
 The [task](../images/task-status.png) and [tree screenshots](../images/agent-tree.png) render the current plugin with official frontend components and sample data. They illustrate the UI, not model quality.
@@ -181,6 +202,7 @@ maxContextBytes: 32768       # Super Code: task-memory context bytes
 ```
 
 - A shorter deadline set explicitly on a call still applies.
+- After disabling or uninstalling the plugin, PTC calls whose execution inputs have not yet been resolved use the host deadline. Already-started calls keep their original deadline and are not restarted to restore settings.
 - Older hosts time PTC by busy CPU time; newer hosts by elapsed time. The two aren't comparable.
 - Raising `maxTasks` or `maxContextBytes` uses more context. The defaults are usually fine.
 
@@ -240,10 +262,13 @@ Under the **Connection** tab:
 
 **Allow caller approval** is off by default. When enabled, an exact one-time tool approval for a delegated task can be sent to the caller through an interactive form or `dsh_approve`. Host policy remains authoritative, and directory consent always stays in DSH.
 
+MCP tasks and connection settings are stored in the DSH profile. Keep this profile on a local filesystem with atomic hard-link support, such as APFS, NTFS or ext4, and do not share it across machines. exFAT and FAT cannot provide this lock mechanism; the startup log explains how to move the profile. This requirement applies to profile storage, not to the working directories you grant access to.
+
 Directories:
 
 - The DSH workspace list is reread every time you open the tab. On first setup, existing workspaces are preselected; workspaces added later are **not** granted access automatically.
 - Add directories with the directory picker. Uncheck to revoke; unchecking the last one also turns MCP off.
+- After revocation, the old directory request reports rejection. The caller must submit a new request to ask for access again.
 
 Connections:
 
@@ -251,7 +276,7 @@ Connections:
 - Disconnect clients from the Connection tab; idle connections drop after 10 minutes. Disconnecting doesn't cancel accepted tasks.
 - "Connected conversations" on Home means the linked MCP connection is still online, regardless of whether the task finished. After disconnect they count as past conversations.
 
-Port conflicts are reported directly. Use the URL shown on the page.
+Port conflicts are reported directly. Choose another port, or resolve the conflict and click **Retry**. If settings were saved but connection cleanup failed, **Retry** restores the connection using those settings. Failed saves retain your draft for retry. Use the URL shown on the page.
 
 ### Configure the client
 
@@ -267,10 +292,10 @@ Whether a client can set up MCP from the prompt by itself depends on the client.
 |---|---|
 | `dsh_status` | Inspect directories, policies, capacity, blocking tasks, and recovery guidance |
 | `dsh_models` | List exact model/effort IDs for `dsh_run` and `dsh_send`; omission retains the current/default selection |
-| `dsh_read` | Page through owned conversations and managed children, including public messages, tool results, and failure details; excludes reasoning and private configuration |
+| `dsh_read` | Page through owned conversations and managed children, including public messages, tool calls/results (also PTC sub-calls), and failure details; excludes reasoning and private configuration |
 | `dsh_approve` | Approve or reject one exact pending operation only when caller approval is enabled |
 | `dsh_sessions` | Call first after connecting: lists recent conversations per directory with status, whether they can be reused, connection capacity, and task concurrency |
-| `dsh_run` | Start a task. Pass `taskId` to reuse a specific conversation, or `newConversation: true` for a new one (passing neither also creates new). It **never** picks the latest conversation on its own. Same `requestId` + same input won't dispatch twice; different input is a conflict |
+| `dsh_run` | Start a task. Pass `taskId` to reuse a specific conversation, or `newConversation: true` for a new one (passing neither also creates new). `newConversation: false` requires `taskId` and never creates or automatically picks the latest conversation. Same `requestId` + same input won't dispatch twice; different input is a conflict |
 | `dsh_wait` | Read task status, output, workspace changes, and new events, waiting up to 30 seconds. Without a task ID, lists all tasks page by page |
 | `dsh_send` | Add a follow-up or continue in the same conversation, with a new `requestId`. Never approves permissions for you |
 | `dsh_request_directory` | Ask for a new working directory; you approve it in DSH |
@@ -280,6 +305,8 @@ Whether a client can set up MCP from the prompt by itself depends on the client.
 
 The client calls `dsh_request_directory` with a `requestId`, an existing absolute `path`, and a `reason`. DSH shows a global dialog (Settings doesn't need to be open) with the client name, directory, and reason.
 
+Before approval, DSH rechecks the requested directory's identity and the request status. Replacing the directory at the same path, expiry, or withdrawal prevents the old request from granting access; use a new `requestId` to request consent again. Ordinary file changes inside that directory do not invalidate consent.
+
 - Access is granted only when you click **Allow access**. A model's auto-approval can't do it for you.
 - Repeat the same call to check status: `pending` / `approved` / `rejected` / `expired`.
 - Tasks can't run in that directory until it's approved.
@@ -295,6 +322,9 @@ The client calls `dsh_request_directory` with a `requestId`, an existing absolut
 - `finished`: execution ended, but `acceptance` is always `not-verified`. The caller must check the result.
 - Returned file info is an observation of the Git workspace, including changes that existed before the task, so it doesn't prove the task made them. Non-Git directories or exceeded limits are marked unavailable.
 - Wait timeouts and client disconnects don't cancel tasks. Turning off the connection just stops new requests.
+- Requests still being prepared recheck authorization before creating a conversation, selecting a model, and sending a prompt. Closing the service, revoking its directory, or disconnecting stops unsent prompts; submitted tasks keep running. Reconnecting never automatically resends a saved intent, and an unsent continuation preserves the previous turn's state.
+- Creation, continuation, and cancellation stop if saving their intent fails. A restart during continuation preparation preserves the previous turn's tracking. An unconfirmed new prompt never inherits completion from the previous turn. Task records share a 64 MiB read/write limit; oversized writes are refused without replacing the saved file.
+- Cancelling an MCP wait or approval request prevents a late form answer from granting permission. Service, directory and caller-approval settings are checked again before the decision. A one-time approval already decided is neither undone nor granted again by a later cancellation.
 - After a restart, prompts are never resent automatically. Find the ID via `dsh_sessions` or the task list, check the state, then continue explicitly. If the state can't be confirmed, it shows `interrupted` or `unavailable`; tasks not confirmed as stopped keep their concurrency slot.
 
 On capacity refusal, retain the same `requestId` and input and wait through `dsh_status` and `dsh_wait`. Do not cancel other clients' tasks or reconfigure the serving app. Release SDK sessions with `transport.terminateSession()` before `client.close()`; closing alone does not release server capacity. The packaged `super-code-mcp` CLI accepts JSON on stdin and reads its token from the private `DSH_MCP_TOKEN` environment variable.
@@ -305,17 +335,63 @@ On capacity refusal, retain the same `requestId` and input and wait through `dsh
 - Limits: 200 task records, 128 follow-ups per task, the latest 200 events, the last 12,000 output characters. Directory-request history keeps up to 100 entries, in memory only.
 - When task records are full, new tasks are rejected explicitly; old records are never dropped silently.
 
+## Preset Skill packages
+
+Open **Settings → Super Code → Preset Skills**. All packages are off by default; existing saved choices are preserved. Individual Skills and workflows have separate groups; workflows remain whole-package switches:
+
+| Category | Name | Purpose | Default |
+|---|---|---|---|
+| Skill | grill-me | Sharpen requirements, plans and decisions through rounds of questions. | Off |
+| Skill | OpenSpec | Organize requirements, design, tasks, implementation and verification around specifications. | Off |
+| Skill | oil-ui | Design, improve and review frontend interfaces. | Off |
+| Workflow | Matt Pocock Skills | Question requirements, plan, develop, test, debug and review code through complete engineering workflows. | Off |
+| Workflow | PStack | Collaborative architecture, development, verification and review workflows. | Off |
+
+The independent grill-me switch reuses Matt Pocock 1.3.2's original `grill-me` entry and its `grilling` dependency without enabling the whole workflow package. The complete Matt Pocock package still includes both; enabling both switches does not register duplicates.
+
+Matt bundles the complete official plugin distribution (1.3.2, 27 Skills), not a selected subset. OpenSpec bundles its complete static skills.sh distribution (12 Skills). PStack is a Cursor plugin containing 58 standard `SKILL.md` entrypoints (0.15.36). All its Skills, references and helpers are bundled; Cursor plugins, agents and automations are not registered. Provenance, inventories and licenses are included in each package's `UPSTREAM.md` and `distribution.json` under `builtin/skills/`.
+
+Changes save automatically. Once accepted, they update the Skill catalog, subsequent loads and native slash menus in existing conversations without refreshing the page, creating a conversation or switching presets. Disabling a package does not retract instructions already loaded into a conversation; start a fresh conversation when you need an empty context. Project/user Skills are unaffected, including same-name overrides. The original invocation rules remain: some Matt Skills require an explicit user invocation instead of automatic model selection.
+
+Enabled packages are copied in full to `skills/dsh-super-code/` under the DSH user home, preserving references and helper paths. A custom `DSH_HOME` is respected. **Open Skill directory** opens `skills/` in the file manager on the device running DSH; it is unavailable on headless hosts. Disabling removes unchanged plugin copies. User edits and existing conflicting directories are preserved and reported; inspect them before retrying sync. Uninstalling the plugin stops registering these nested resources.
+
+In a main conversation using the Super Code preset, type `/` to select a Skill from DSH’s native menu, or enter `/oil-ui`, `/grill-me`, `/openspec-propose` or `/poteto-mode` followed by your task. Disabled packages no longer supply commands or bodies. Search by package with `/pstack` or `/mattpocock` to find enabled Skills, then pick one to insert its real `/skill-name`. Package names are search terms, not standalone commands. The plugin uses public DSH APIs to supplement package search and refresh the native menu without restarting the host or changing drafts or model calls. In-flight model requests and instructions already in context remain.
+
+Native same-name precedence is preserved: project/user Skills take priority. With Matt and PStack both enabled, `/tdd` and `/teach` use the earlier-registered Matt version; disable Matt to use PStack’s versions.
+
+PStack uses a converted edition for **DSH `0.2.1-alpha.1` and later**, retaining all 58 entrypoints and 23 playbooks. The adapter handles delegation, model selection, history access, Skill paths and long-run checkpoints. `/setup-pstack` writes only optional project `.dsh/pstack.md` preferences; `/make-bot-ui` uses the plugin's MCP connection. Models inherit the host selection by default and the root schedules dependent phases. Without history tools, workflows use explicit handoffs or project artifacts rather than private host records. Original upstream files remain for upgrade comparison; no global prompt is changed.
+
+Configuration retains all 17 upstream role categories. Reasoning budget, panel size and concurrency are distinct; when per-call model or effort selection is unavailable, the workflow explicitly inherits the host. Reflect retains three reviews followed by independent synthesis. A host without timed resumption can preserve a checkpoint but cannot promise scheduled work after the turn ends.
+
+Browser, Git/GitHub and model-checking workflows still require real tools. Helpers may need Bun, authenticated gh, mise/Java/TLC or Pele, subject to the environment and user authorization. Enabling the package runs no helpers or installs. Missing capabilities are reported; sequential self-review is not presented as independent or cross-model verification.
+
+The plugin manages directory synchronization globally; reinstalling a preset is unnecessary. A preset must still expose the host's native Skill tools to load and invoke Skills in that conversation. The Super Code preset includes them.
+
+Each bundled package displays its own pinned version. The matching OpenSpec CLI is installed with the plugin and exposed as `openspec` inside enabled Super Code shell calls, with no global installation. Switching, listing and loading Skills execute no commands and change no project files, dependencies or configuration. Run `openspec init` only after an explicit user request. Matt's tracker workflows may require an authenticated `gh`/`glab`; its setup workflow runs only when explicitly requested. Package switches do not grant permission for commands, commits, remote writes, or paid services. Skill resources update with this plugin.
+
 ## Frontend Skill
 
-The plugin ships `super-code-frontend`, used for pages, components, visual tweaks, responsive layout, interaction states, and browser checks. It focuses on:
+The plugin bundles [oil-ui](https://github.com/oil-oil/oil-ui) **0.32.0**, replacing `super-code-frontend` for designing, improving, and reviewing pages, apps, dashboards, and components. It covers:
 
-- whether elements line up on the same edges
-- whether what you see matches what you can click
-- whether controls stay put, are easy to reach, and need little input
+- User tasks, design direction, visual hierarchy, and key interactions
+- On-demand style comparison pages with local assets for offline viewing
+- Reviewing and refining UI using rendered screenshots and interaction results
 
-It keeps the project's own visual style, needs no particular framework, and adds no tools. Normally only its name and short description go to the model; the body and relevant references load when a task matches. You can also ask for it directly: "use super-code-frontend".
+Enable oil-ui in Preset Skills settings first; it then loads through Harness's native Skill mechanism and follows the project's framework and implementation conventions. Normally only its name and short description go to the model; the body and relevant references load when a task matches. You can also ask for it directly: "use oil-ui".
 
-- Only presets installed or reinstalled from 0.3.3 onward include it. For existing presets, back up your customizations, then reinstall.
+- Existing directory presets are not overwritten automatically. After updating to a plugin containing oil-ui, back up customizations and reinstall the preset from Settings. Declarative presets read the resources directly from the current plugin package.
 - A project or user skill with the same name overrides the bundled one, following host precedence.
-- The browser-check part is guidance only: no enforced loop detection, no automatic long-term memory writes.
+- This bundled copy updates with the plugin; it does not check for updates separately or run the upstream installer. See [provenance, adaptations, and licenses](../../builtin/skills/oil-ui/UPSTREAM.md).
+- The design guidance needs no extra dependencies. The optional comparison generator requires Python 3.10+; screenshot and image-processing scripts need Node.js 22+ and Chrome, Chromium, or Edge. Some media operations also need ffmpeg/ffprobe. Host tools can be used when these are unavailable, with unverified behavior stated explicitly.
+- Optional media searches contact their providers. Oil image generation needs an account and explicit authorization to spend credits; plugin startup never invokes it. The upstream Pro recommendation script writes a one-time local marker only when used as directed by the Skill.
 - Whether it improves frontend quality needs a separate real-task evaluation; install tests and SWE scores can't show that.
+
+## Direct calls and action-local help
+
+Both state tools support `{"action":"help","forAction":"read"}`. One help response includes fields, an example, outputs and recovery rules without accessing state or retrying persistence. Replace example sources/identities/versions with observed values.
+
+`super_code_task {"action":"read"}` reads the focused task; no focus is an explicit error, never a guess. Updates still require an observed expectedRevision. Shape validation precedes task barriers; errors include repair examples. No automatic alias/type/source/revision correction is performed.
+
+`super_code_memory {"action":"read","topic":"build"}` reads a project topic without creating a task or changing focus/context injection. Omitted topic retains the focused-topic default. Explicit reads do not borrow another task's code version. details retains scope/topic; add id to read the selected entry. topic is accepted only for project read. Writes retain their existing focus, provenance and concurrency rules. All memory is historical.
+
+Input rejection appends no new business change; checkpoint failure may follow an appended event. Confirm using list/read instead of replaying writes. Event contracts and tool names are unchanged. Local verification does not establish model quality or token savings.

package/docs/images/npm-downloads-dark.svg

@@ -0,0 +1,22 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="960" height="320" viewBox="0 0 960 320" role="img" aria-labelledby="title description">
+<title id="title">dsh-super-code npm downloads</title>
+<desc id="description">dsh-super-code · weekly · last 13 weeks. 1,225 last 7 days. Through 2026-11-11. Source: npm.</desc>
+<style>text{font-family:system-ui,-apple-system,BlinkMacSystemFont,'Segoe UI',sans-serif;font-size:12px;fill:#9da7b3;font-variant-numeric:tabular-nums}</style>
+<rect x=".5" y=".5" width="959" height="319" rx="12" fill="#0d1117" stroke="#30363d"/>
+<text x="28" y="35" style="font-size:19px;font-weight:600;fill:#e6edf3">npm downloads</text>
+<text x="28" y="58">dsh-super-code · weekly · last 13 weeks</text>
+<text x="932" y="35" text-anchor="end" style="font-size:19px;font-weight:600;fill:#e6edf3">1,225 last 7 days</text>
+<text x="932" y="58" text-anchor="end">Through 2026-11-11</text>
+<path d="M68 260H926" stroke="#212830"/><text x="56" y="264" text-anchor="end">0</text>
+<path d="M68 201.67H926" stroke="#212830"/><text x="56" y="205.67" text-anchor="end">500</text>
+<path d="M68 143.33H926" stroke="#212830"/><text x="56" y="147.33" text-anchor="end">1,000</text>
+<path d="M68 85H926" stroke="#212830"/><text x="56" y="89" text-anchor="end">1,500</text>
+<polyline points="68,260 139.5,260 211,260 282.5,260 354,260 425.5,168.07 497,187.08 568.5,209.83 640,119.53 711.5,195.37 783,118.48 854.5,116.15 926,117.08" fill="none" stroke="#e05d58" stroke-width="2.5" stroke-linejoin="round" stroke-linecap="round"/>
+<circle cx="926" cy="117.08" r="3.5" fill="#e05d58"/>
+<text x="68" y="285" text-anchor="middle">08/19</text>
+<text x="282.5" y="285" text-anchor="middle">09/09</text>
+<text x="497" y="285" text-anchor="middle">09/30</text>
+<text x="711.5" y="285" text-anchor="middle">10/21</text>
+<text x="926" y="285" text-anchor="middle">11/11</text>
+<text x="28" y="307">Source: npm</text>
+</svg>

package/docs/images/npm-downloads.svg

@@ -0,0 +1,22 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="960" height="320" viewBox="0 0 960 320" role="img" aria-labelledby="title description">
+<title id="title">dsh-super-code npm downloads</title>
+<desc id="description">dsh-super-code · weekly · last 13 weeks. 1,225 last 7 days. Through 2026-11-11. Source: npm.</desc>
+<style>text{font-family:system-ui,-apple-system,BlinkMacSystemFont,'Segoe UI',sans-serif;font-size:12px;fill:#59636e;font-variant-numeric:tabular-nums}</style>
+<rect x=".5" y=".5" width="959" height="319" rx="12" fill="#ffffff" stroke="#d8dee4"/>
+<text x="28" y="35" style="font-size:19px;font-weight:600;fill:#1f2328">npm downloads</text>
+<text x="28" y="58">dsh-super-code · weekly · last 13 weeks</text>
+<text x="932" y="35" text-anchor="end" style="font-size:19px;font-weight:600;fill:#1f2328">1,225 last 7 days</text>
+<text x="932" y="58" text-anchor="end">Through 2026-11-11</text>
+<path d="M68 260H926" stroke="#e8ecf0"/><text x="56" y="264" text-anchor="end">0</text>
+<path d="M68 201.67H926" stroke="#e8ecf0"/><text x="56" y="205.67" text-anchor="end">500</text>
+<path d="M68 143.33H926" stroke="#e8ecf0"/><text x="56" y="147.33" text-anchor="end">1,000</text>
+<path d="M68 85H926" stroke="#e8ecf0"/><text x="56" y="89" text-anchor="end">1,500</text>
+<polyline points="68,260 139.5,260 211,260 282.5,260 354,260 425.5,168.07 497,187.08 568.5,209.83 640,119.53 711.5,195.37 783,118.48 854.5,116.15 926,117.08" fill="none" stroke="#e05d58" stroke-width="2.5" stroke-linejoin="round" stroke-linecap="round"/>
+<circle cx="926" cy="117.08" r="3.5" fill="#e05d58"/>
+<text x="68" y="285" text-anchor="middle">08/19</text>
+<text x="282.5" y="285" text-anchor="middle">09/09</text>
+<text x="497" y="285" text-anchor="middle">09/30</text>
+<text x="711.5" y="285" text-anchor="middle">10/21</text>
+<text x="926" y="285" text-anchor="middle">11/11</text>
+<text x="28" y="307">Source: npm</text>
+</svg>