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
Notable: A new repo — autoresearch — was created during this window (2026-03-19T17:44Z). All 14 opened/closed issues are automated workflow reports from zircote/.github (smart-alerts, maintenance-board, daily standup/triage/nudge).
Active Repositories
autoresearch 🆕
Autonomous skill improvement loop for Claude Code plugins — inspired by Karpathy's autoresearch.
Created: 2026-03-19T17:44Z
Commits (5+ to main)
6d997dbd fix: align test fixtures with eval expectations — @zircote
4dd92ecc perf: harden autoresearch with deterministic evals and execution directives — @zircote
07c55929 perf: integrate deterministic checks into autoresearch loop — @zircote
Full list: 23 PRs total. Most are stale Dependabot bumps across the plugin ecosystem repos (atlatl-spec, subcog, rlm-rs, daedalus, and others).
Failing CI on Default Branch
Repo
Workflow
Age
Root Cause
atlatl
Pipeline + Code Coverage
~19h
refactor: replace std::sync::Mutex with tokio — compilation failure
atlatl-spec
Validate Specification
~13 days
Invalid <br/> in Mermaid sequence diagram
atlatl-spec
Deploy to GitHub Pages
~19 days
@redocly/cli 2.20.0→2.20.4 breaking change
vscode-git-adr
CI (dependabot PR)
~23 days
upload-artifact v7 migration needed
sdlc-quality CI failures also noted in recent smart-alerts (Dependabot config regression).
Security-Related Activity
None in the last 24 hours.
Theme of the Day: Autoresearch Rollout 🔬
Today's activity reflects a coordinated eval-driven development push across the plugin ecosystem. @zircote launched the new autoresearch repo and added deterministic + LLM eval suites across 9 repositories (documentation-review, github-social, aesth, version-guard, adr, mcp-bundle, claude-team-orchestration, refactor). This represents the Karpathy-inspired autonomous skill improvement loop being operationalized at scale.
Quiet Repositories
67 repositories with no activity in the last 24 hours
swagger-php, Hal, chef-composer, rlm-rs, subcog, Bloom, Rhubarb, ApiProblem, mnemonic, atlatl, atlatl-spec, vscode-git-adr, sdlc-quality, github-project-manager, daedalus, MIF, git-notes-memory, zircote-ansible, csharp-lsp, python-lsp, lsp-tools, gh, adrscope, nsip, rust-template, and ~42 additional plugin and infrastructure repositories with updated_at prior to the 24-hour threshold.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
Monitored: 77 repositories | Excluded: php-swagger
Organization Totals
Active Repositories
autoresearch🆕Commits (5+ to
main)6d997dbdfix: align test fixtures with eval expectations —@zircote4dd92eccperf: harden autoresearch with deterministic evals and execution directives —@zircote07c55929perf: integrate deterministic checks into autoresearch loop —@zircote0ebe1961feat: add deterministic checker for zero-variance eval grading —@zircote5aa2394fstyle: fix Diataxis compass box alignment in docs index —@zircotedocumentation-reviewCommits (5 to
main)49e8b867test: add eval workspace with 3 iterations of benchmark results —@zircote60125d04chore: update commands, templates, docs, and remove legacy agent files —@zircote0549c7fafeat: add eval test suites for all 4 skills —@zircotec2e0ba1cperf: improve skills with eval-driven quality standards —@zircote5cb92353feat: add agent directories, manifest, language toolchains reference —@zircotemcp-bundleCommits (2 to
main)902cd0e1feat: add deterministic + LLM eval suite for mcpb skill —@zircote2456519bchore: ignore autoresearch workspace directories —@zircotegithub-socialCommits (1 to
main)bff9bcecfeat: add autoresearch eval suites for all 4 skills —@zircoteaesthCommits (2 to
main)c3229c83feat: add autoresearch evals for aesth skill —@zircotee0f3e705perf: strengthen aesth design principles enforcement —@zircoteadrCommits (2 to
main)b061c0b7perf(skills): improve adr-integration with working code examples —@zircotef33ef917feat(skills): add eval suites for all 12 ADR skills —@zircoteversion-guardCommits (2 to
main)b42d9403perf: improve context7 query patterns for version extraction —@zircote7acf1996feat: add autoresearch evals for version-guard skill —@zircoterefactorCommits (3 to
main)91541b42chore: remove diataxis audit report —@zircotee7e3a25fdocs: fix all diataxis audit findings across existing docs —@zircotefd2af0f6feat: add feature-dev tutorial, how-to guide, and diataxis audit —@zircoteclaude-team-orchestrationCommits (2 to
main)99165443chore: add gitignore for autoresearch workspaces —@zircotefd54e7d9feat: add autoresearch evals for all 10 skills —@zircote.githubIssues (automated workflow reports only)
Attention Required
Stale PRs (> 7 days, no activity)
23 open PRs not updated since before 2026-03-13:
git-notes-memory@zircote(draft)zircote-ansible@conradoFailing CI on Default Branch
atlatlrefactor: replace std::sync::Mutex with tokio— compilation failureatlatl-spec<br/>in Mermaid sequence diagramatlatl-spec@redocly/cli2.20.0→2.20.4 breaking changevscode-git-adrupload-artifactv7 migration neededSecurity-Related Activity
None in the last 24 hours.
Theme of the Day: Autoresearch Rollout 🔬
Today's activity reflects a coordinated eval-driven development push across the plugin ecosystem.
@zircotelaunched the newautoresearchrepo and added deterministic + LLM eval suites across 9 repositories (documentation-review,github-social,aesth,version-guard,adr,mcp-bundle,claude-team-orchestration,refactor). This represents the Karpathy-inspired autonomous skill improvement loop being operationalized at scale.Quiet Repositories
67 repositories with no activity in the last 24 hours
swagger-php, Hal, chef-composer, rlm-rs, subcog, Bloom, Rhubarb, ApiProblem, mnemonic, atlatl, atlatl-spec, vscode-git-adr, sdlc-quality, github-project-manager, daedalus, MIF, git-notes-memory, zircote-ansible, csharp-lsp, python-lsp, lsp-tools, gh, adrscope, nsip, rust-template, and ~42 additional plugin and infrastructure repositories with
updated_atprior to the 24-hour threshold.Generated by org-monitor workflow — https://github.com/zircote/.github/actions/runs/23334239129
Beta Was this translation helpful? Give feedback.
All reactions