Skip to content

Add new Everforest theme - #77

Closed
mono-chrome wants to merge 3 commits into
anomalyco:devfrom
mono-chrome:feature/add-theme-everforest
Closed

Add new Everforest theme#77
mono-chrome wants to merge 3 commits into
anomalyco:devfrom
mono-chrome:feature/add-theme-everforest

Conversation

@mono-chrome

Copy link
Copy Markdown
Contributor

This adds the Everforest theme to opencode.
Theme reference: https://github.com/sainnhe/everforest

Additionally updates README.md and cmd/schema/main.go with theme references.

@mono-chromemono-chrome changed the title Add new Everforest theme.Add new Everforest themeJun 10, 2025
@thdxr

Copy link
Copy Markdown
Member

closing as we rewrote the codebase - appreciate it though!

@thdxrthdxr closed this Jun 14, 2025
@mono-chromemono-chrome mentioned this pull request Jun 17, 2025
randomm referenced this pull request in randomm/opencode Jan 9, 2026
)
* fix: eliminate remory syntax confusion and consolidate memory files (#74)
- Consolidated memory-protocol-mandatory.md and remory-cli-reference.md into new memory-and-remory.md
- Fixed contradictions in research-specialist.txt prompt
- Updated configuration files to reference consolidated memory documentation
- Reduces duplication and improves clarity of remory syntax and memory protocols
* fix(#76): enforce git-agent scope boundaries to prevent code fixing
- Add explicit bash command denylist to both config files for critical operations
- Add hard-stop boundaries and VIOLATION DETECTION patterns to git-agent.txt
- Add PRE-ACTION CHECKS validation before every operation
- Enhance linting discovery to READ-ONLY operations only
- Add enhanced CI failure delegation protocol with mandatory specialist routing
These changes prevent git-agent from attempting to fix code issues, ensuring proper
delegation to specialized agents (rust-tdd-architect, python-best-practices, etc.)
andreipromarketing-dev pushed a commit to andreipromarketing-dev/opencode that referenced this pull request Apr 7, 2026
* fix: use correct unscoped agent-browser package name
The @anthropic-ai/agent-browser package does not exist on npm.
The correct package is the unscoped 'agent-browser' from Vercel Labs (v0.7.6).
Fixes the issue noted in PR anomalyco#75 review comment.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* fix: correct agent-browser API to match actual package exports
- Change AgentBrowser to BrowserManager (correct export name)
- Update documentation to show actual CLI-based usage pattern
- Add snapshot + refs system examples (the real agent-browser API)
- Add programmatic BrowserManager API for advanced use cases
- Show CLI wrapper approach for script integration
The agent-browser package from Vercel is primarily a CLI tool,
not a JavaScript library with semantic methods. This corrects
the documentation to match the actual package API.
Fixes CodeRabbit review comment about incorrect import/export.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
bussard76 pushed a commit to bussard76/openwork that referenced this pull request May 12, 2026
mliotta added a commit to mliotta/opencode that referenced this pull request Aug 5, 2026
Adds opt-in integration tests (`bun run test:car`) covering the paths that only
break against a real daemon: delegated inference streaming end to end,
concurrent calls sharing one runner connection without cross-talk, host abort
cancelling a delegated call, and facts being written and recalled. Every defect
found in this integration so far was a wire-contract defect (car-releases#75,
anomalyco#76, anomalyco#77) that a mock would not have reproduced.
Isolation is the point. CAR sandboxes memory to $HOME/.car/memory, so running
these against a developer's daemon writes real memory for temp projects that
never existed — an earlier run left ~36 MB of orphaned facts in a developer's
CAR state. Each run gets a throwaway car-server with its own HOME, OS-assigned
port, agents manifest and auth token, and the temp home is removed on teardown.
The daemon has to be started by a parent script rather than in beforeAll:
car-runtime's native binding reads CAR_DAEMON_URL from the C environ, and bun
does not propagate runtime process.env mutations to it. Setting it in a
beforeAll looks correct and silently does nothing — the client then connects to
whatever daemon the developer already has running, which is how the first
version of these tests failed with "model not found" against a real daemon.
Verified separately that CAR_DAEMON_URL is honoured when set at spawn time.
Skipped unless OPENCODE_CAR_INTEGRATION=1 and the car-server binary is present,
so the default suite stays fast, green and daemon-free.
mliotta added a commit to mliotta/opencode that referenced this pull request Aug 6, 2026
CarHost auto-updates via Sparkle and had already moved to 0.46.0, leaving
opencode's pinned 0.44.0 client older than the daemon it talks to — the reverse
of the skew the version rule exists to prevent. car-runtime is now excluded from
bunfig's minimumReleaseAge so the client can track a daemon that updates itself;
the tradeoff is accepted deliberately for this one package, which we publish.
Correlate delegated calls with GenerateRequest.client_ref (car-releases#78). The
UUID was previously smuggled through `prompt`, the only field surviving the
serde round-trip, which held only because delegated models ignore it. `prompt`
is still a required field, so it stays as an empty placeholder — the integration
suite caught that when it was dropped entirely.
Verified against a real 0.46.0 daemon rather than by inspection:
- client_ref is echoed verbatim into inference.runner.invoke, with prompt set to
a throwaway value to prove it is no longer load-bearing (anomalyco#78)
- completing with no preceding event yields exactly one invoke and preserves
RunnerResult.text; previously two invokes and an empty result (anomalyco#76)
- concurrent delegated calls no longer cross-talk (anomalyco#77)
- car-releases#75 is improved but open: 5066/5108/5289ms on 0.44.0 became
1879/4181/4251ms on 0.46.0 — a constant became variable seconds, so CAR stays
off by default
Not adopting per-project memory namespaces yet, despite 0.46.0 adding them.
Namespace memory is never written to disk: it is lost on daemon restart, and
memory.persist silently writes zero records while a namespace is bound
(car-releases#82). Since CarHost restarts on every update, binding one today
would trade the cross-project leakage of anomalyco#79 for silent data loss. Staying on
the shared graph, with the tradeoff documented, until that closes.
CAR integration suite: 11 pass against a live 0.46.0 daemon, and it now runs in
half the wall time. CAR off: 3231 pass, 1 fail — a load flake that passes 21/21
in isolation.
murtaza64 added a commit to murtaza64/opencode that referenced this pull request Aug 12, 2026
Agents cite tickets bare (SD-123, dotfiles#77, anomalyco#77) with no context. Wrap
refs found in rendered transcript text (idempotent MutationObserver pass
alongside stampTools) and show a hover card: tracker-backed refs fetch
issue info via the es-dashboard issue API (title, state, claims/labels,
assignees, body snippet) with links to the in-app ticket page and the
external tracker; jira-style and owner/repo#N refs get direct links.
Card repositions with transcript auto-pin scrolls and closes when the
anchor leaves the viewport.
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants

@mono-chrome@thdxr@adamdotdevin@jpoz