[jonbeckman]

Summary

Adds agent skills for the tools this repository uses that were not recorded in skills-lock.json.

• npx skills update -p -y found the other locked skills current. The Cloudflare skill update also deletes .agents/skills/cloudflare/references/c3/ (README.md, api.md, configuration.md, gotchas.md, patterns.md). Those files stay, and that skill update is not included.
• Oxlint and oxfmt skills come from delexw/claude-code-misc (957 and 41 installs). They are usage guides for projects that already run these tools. The official Oxc skills are migration guides.
• Tegami has one published skill, migrate-from-changesets (1 install). This repository already uses Tegami, so that migration skill is not installed. .agents/skills/tegami points at scripts/tegami.mts, docs/releasing.md, and the manage-solzero-releases skill.
• .oxfmtrc.json and oxlint.config.ts ignore .agents/skills/** and .claude/skills/**. nub run format:fix left all 802 skill files byte-identical.
• Alchemy stays on wyattjoh/skills. The e2e skill is unchanged. No package versions, VERSION, or nub.lock changes.

release:none: this is agent guidance only. Users, administrators, and deployment operators see no product change, so there is no .tegami/ entry.

Leave this pull request open. A push to master publishes GHCR images.

Validation

• nub run config:check passed.
• nub run typecheck passed.
• nub run lint passed.
• nub run format passed. nub run format:fix did not change any skill file.
• nub run test:config passed (31 tests).
• nub run test did not start. Docker is not installed in this environment. The diff is agent skill text and ignore patterns.

Checklist

• I added or updated tests for behavior changes, or explained why tests are not needed. Skill text does not change application behavior.
• I updated documentation and config examples when behavior or configuration changed. No product behavior or configuration changed.
• I added a .tegami/ release entry, or explained why this pull request uses release:none.
• I ran nub run config:check, nub run typecheck, nub run lint, nub run format, and nub run test. The full e2e suite could not start without Docker; the account-free config suite passed.
• I did not include credentials, private data, internal URLs, or organization-specific defaults.
• I updated THIRD_PARTY_NOTICES.md for new third-party code or separately licensed assets. Agent skills stay recorded in skills-lock.json, matching earlier skill refreshes. They are not shipped product code.

  

[github-actions]

🟢 e2e solzero-core: no tests selected, 30 left out by filters

Change 6 files changed since c7553a4
Filter decision left out all 30 tests (1 call, 16.1k tokens)

30 tests left out by filters

Test	Filter	Reason
tests/accounts.e2e.ts › create, use and revoke an isolated user API key	decision	5% likely affected
tests/accounts.e2e.ts › Better Auth 1.7 account APIs select the local account row id	decision	5% likely affected
tests/accounts.e2e.ts › Accounts settings shows GitHub as unlinked without OAuth providers	decision	5% likely affected
tests/admin.e2e.ts › administrator pages show live control-plane configuration	decision	5% likely affected
tests/advanced-sessions.e2e.ts › incognito sessions stay outside normal lists but open directly	decision	5% likely affected
tests/advanced-sessions.e2e.ts › session MCP definitions and subagent preferences persist	decision	5% likely affected
tests/api-contract.e2e.ts › exported API key and session schemas decode actual Worker responses	decision	5% likely affected
tests/auth.e2e.ts › sign-in gate protects every application route	decision	6% likely affected
tests/auth.e2e.ts › credential sign-in opens the requested settings page	decision	3% likely affected
tests/auth.e2e.ts › anonymous callers cannot access any protected API group	decision	2% likely affected
tests/auth.e2e.ts › sign-out removes authenticated UI access	decision	2% likely affected
tests/bots.e2e.ts › create a bot and manage a temporary routine through the UI	decision	5% likely affected
tests/branding.e2e.ts › the selected deployment brand reaches the browser	decision	2% likely affected
tests/branding.e2e.ts › health identifies the configured application without exposing configuration	decision	3% likely affected
tests/branding.e2e.ts › OpenAPI uses the selected deployment brand	decision	3% likely affected
tests/session-transfer.e2e.ts › admin authentication transfer redeems once and retains its deep link	decision	5% likely affected
tests/sessions.e2e.ts › session tools, archive and restoration persist through the real Worker	decision	5% likely affected
tests/sessions.e2e.ts › a second account cannot read or mutate another account's session	decision	5% likely affected
tests/sessions.e2e.ts › isolate executes a real prompt and persists its output	decision	8% likely affected
tests/sessions.e2e.ts › opencode executes a real prompt and persists its output	decision	7% likely affected
tests/sessions.e2e.ts › codex executes a real prompt and persists its output	decision	5% likely affected
tests/settings.e2e.ts › all settings categories load real data	decision	5% likely affected
tests/settings.e2e.ts › runtime default persists after reload	decision	5% likely affected
tests/settings.e2e.ts › MCP deep link and legacy settings URL open the MCP tab	decision	5% likely affected
tests/settings.e2e.ts › secret editor creates, finds and deletes a disposable secret	decision	5% likely affected
tests/settings.e2e.ts › theme preference survives reload	decision	5% likely affected
tests/skills.e2e.ts › global skill defaults and per-user preferences persist independently	decision	8% likely affected
tests/workflows.e2e.ts › workflow authoring, rename, export, disable and archive persist	decision	5% likely affected
tests/workflows.e2e.ts › manual workflow executes JavaScript and stores a real R2 artifact	decision	5% likely affected
tests/workflows.e2e.ts › invalid workflow definitions are rejected before storage	decision	5% likely affected

_{e2e 0.21.3 · 15ms · local-cloudflare · run artifacts}

[github-actions]

Tegami

This repository uses Tegami to manage releases. When your changes affect published packages, add a changelog file under .tegami/ before merging.

Create a changelog → · Changelog format

Release preview

Package	Bump	Version
solzero	major	2.0.0 → 3.0.0

This PR does not add changelog files. Pending changelogs from other branches are included in the preview above.

Run nub run tegami locally to create a changelog interactively.

_{Managed by Tegami.}