You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Focused sync of ARIS Workflow M from upstream wanshuiyin/Auto-claude-code-research-in-sleep at f1bd907b58f653131ebe6807c482e2554e07f9b9 (not a wholesale upgrade of all bundled skills).
Expose /aris-meta-optimize in the shared Auto Research Hub / Chat workflow catalog.
Separate the corpus-read-only proposal producer from manual /aris-meta-apply, with fresh cross-family review, backups, provenance and rollback instructions.
Bundle the required logging, capture, trace, provenance and trigger-evaluation helpers, scoped shared references, licenses and upstream source-hash inventory.
Provide opt-in project logging configuration: preview by default, explicit --apply, atomic backed-up merge preserving unrelated settings/hooks. Global logging is a separate opt-in; standard setup does not enable it.
Use the bundled Codex-exec reviewer bridge in the actual Hub Configure endpoint as well as setup and copied commands. Preserve existing registrations. Reject missing bridge assets rather than registering a dead command.
Harden bridge continuation: validate thread IDs, fail closed on absent/corrupt saved settings, default to read-only and preserve the stored sandbox over config overrides. MCP disconnect cancels active reviewers with bounded TERM/KILL cleanup.
Require full traces for landing even when ARIS_TRACE_MODE=off; require artifact readback before mutation.
Validation
HOME=/tmp/dr-claw-tests-home npm test: 245 tests passed on this ARIS-only branch.
Adds 27 deterministic tests: 23 Workflow M/catalog/helper/bridge tests and 4 mocked actual configure-route tests. Includes a synthetic Python subprocess that ignores TERM to test bounded cleanup (no model/CLI). Temporary HOME/projects only; registration and filesystem writes in route tests are mocked.
Independent review identified the missing actual configure-route update and inherited trace-off behavior; these and a subsequent MCP-disconnect lifecycle finding were fixed and regression-tested.
Safety and limits
No live workflow, model/reviewer invocation, paid inference, global CLI installation, MCP registration or real user settings change was performed during validation.
Logging remains opt-in and may contain sensitive paths/prompt snippets; documentation covers disabling and retention.
Human-only frontmatter / prompts and the optional regex write guard are not an OS sandbox. Dr. Claw Markdown expansion may not enforce every frontmatter permission; documentation states these limitations.
Reviewer model/effort names are requested capabilities with documented error-gated fallback, not a guarantee of account availability.
Shared skill symlinks can affect multiple projects; landing requires explicit acknowledgement of the resolved target scope.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Closes #153.
Focused sync of ARIS Workflow M from upstream
wanshuiyin/Auto-claude-code-research-in-sleepatf1bd907b58f653131ebe6807c482e2554e07f9b9(not a wholesale upgrade of all bundled skills)./aris-meta-optimizein the shared Auto Research Hub / Chat workflow catalog./aris-meta-apply, with fresh cross-family review, backups, provenance and rollback instructions.--apply, atomic backed-up merge preserving unrelated settings/hooks. Global logging is a separate opt-in; standard setup does not enable it.ARIS_TRACE_MODE=off; require artifact readback before mutation.Validation
HOME=/tmp/dr-claw-tests-home npm test: 245 tests passed on this ARIS-only branch.npm run typecheck: passed.npm run build: passed.git diff --check: passed.Safety and limits