The panel stored notes, todos and plans inside one shared JSON file that
six unrelated domains also wrote to, synchronised itself through window
CustomEvents, and could only read plans. It is now Project knowledge:
server-owned storage with explicit routes, a store with rollback, a
section sidebar, plans that open and edit in place, and search across
all of it.
Notes and plans the user pins travel with every message sent in that
project. Pinning is project state, not an attachment to one message, so
it holds until unpinned and the work status panel names what is riding
along and can detach it.
Agent memory is added alongside, in two scopes: what is true about the
user, and what is true about this codebase. The split is not cosmetic —
a wrong project fact costs one project and is noticed, while a wrong
global fact quietly shapes every session everywhere and the user has no
code to check it against. It stays separate from notes so an agent
mistake cannot land in what the user wrote. Sessions receive an index of
titles only; bodies are read on demand, because an index carrying full
text grows until it crowds out the conversation.
Deciding what a session must be told, and whether it has been told, now
lives on the server. The client owned it before, which meant sessions
started without a UI — scheduled tasks, sessions the agent dispatches —
received nothing at all, and a tab's record of what it had sent outlived
the conversation: after compaction the agent no longer held the block
while the tab went on believing it did. What was delivered is recorded
in the session's own metadata, and compaction restores it through the
runtime that already restores pinned messages, in the same turn.
Agent memory ships dark behind OPENCHAMBER_MEMORY_ENABLE: unset, there
is no tool, no routes, no session index, no settings row and no panel
tab. Absent rather than switched off, so nothing invites turning on a
feature that has not been announced. Pinned notes and plans are
unaffected and ship as normal.
* fix(proxy): reuse upstream connections for OpenCode API requests
`createProxyMiddleware` was constructed without an `agent`, so `http-proxy`
fell back to `agent: false`. That disables connection pooling and forces
`Connection: close` on every proxied request, consuming one ephemeral port
per request.
Measured against a real `opencode serve` instance, 200 sequential requests
through the proxy created 201 TIME_WAIT entries (1.005 ports/request). With
a keep-alive agent the same load creates 0.
On macOS the ephemeral range is 16,384 ports and TIME_WAIT lasts 30s, so
sustained traffic around 546 req/sec exhausts the pool — after which every
process on the host fails to open outbound connections with EADDRNOTAVAIL.
`maxSockets: Infinity` preserves the unbounded concurrency of `agent: false`,
so this changes connection reuse only, not request throughput.
Partially addresses #2915.
* fix(proxy): derive proxy agent class from the target scheme
Addresses review feedback on #2916. The first commit created an
unconditional `http.Agent`, which regresses external OpenCode servers
configured over https via `OPENCODE_HOST` (accepted by env-config.js).
http-proxy dispatches through `https.request` when the target protocol is
`https:` (http-proxy/lib/http-proxy/passes/web-incoming.js:126), and
`http.Agent#createConnection` is plain `net.createConnection` — so an
http.Agent would open a plaintext socket to a TLS port and fail every
proxied request. `agent: false` previously worked for both schemes.
`createOpenCodeProxyAgent(target)` now returns an `https.Agent` for https
targets and an `http.Agent` otherwise, derived once from
`resolveProxyTarget()` at registration so the single shared instance is
preserved across `apiProxy` and `interactiveOAuthProxy`.
Guarded in both test layers, verified to fail when the selection is
reverted to an unconditional http.Agent. `https.Agent` extends
`http.Agent`, so the http cases assert `not.toBeInstanceOf(https.Agent)`.
* Round 2: fix: resolve the proxy agent lazily so cold starts honor https
Addresses the round-2 blocker on #2916. Deriving the agent class at
registration is too early: startup-pipeline-runtime.js calls setupProxy()
(line 104) before bootstrapOpenCodeAtStartup() (line 141), so on a fresh
process state.openCodePort is null, buildOpenCodeUrl() throws
(network-runtime.js:86-88), and resolveProxyTarget() returns the http
loopback fallback. An external server configured via OPENCODE_HOST=https://
only appears on state.openCodeBaseUrl after bootstrap, so it was still
getting a plain http.Agent — the regression the previous commit intended
to fix.
`agent` is now a getter backed by a per-scheme memoizing resolver.
http-proxy-middleware rebuilds per-request options with
`Object.assign({}, this.proxyOptions)` in prepareProxyRequest, which invokes
getters, so resolution happens at request time while still yielding one
shared pool per scheme.
Tests now model the production ordering — registration while the port is
null and buildOpenCodeUrl throws, then an https base URL appearing after
bootstrap — and fail against the eager implementation. A behavioral test
pins the http-proxy-middleware option re-read the fix depends on, so a
library change that froze options would fail loudly instead of silently
regressing https targets.
The resolver is module-private; `bun run dead-code` flagged it as an
unused export when it was exported.
* Round 3: docs(changelog): note upstream connection reuse under [Unreleased]
Repo precedent adds [Unreleased] bullets for comparable proxy/stability
fixes (1.18.4 Stability, 1.9.3 Reliability/Proxy). Non-blocker raised in
review on #2916.
* Round 3: docs(changelog): use repo-standard 'behavior' spelling
* Round 4: docs(changelog): don't imply a restart is the only recovery
The ephemeral port pool drains on its own once the exhausting traffic
stops (TIME_WAIT expiry), so a restart is sufficient but not necessary.
Optional nit raised in review on #2916.
* Round 5: fix: construct the proxy agent through one factory; widen the pool
Review found the https branch was mutation-uncovered: the resolver
re-implemented agent construction inline instead of calling the exported
`createOpenCodeProxyAgent(target)`, so replacing its https branch with
`new https.Agent()` — dropping OPENCODE_AGENT_OPTIONS, and with it
keep-alive — left the entire suite green. Since `createOpenCodeProxyAgent`
also had no production callers, its four tests were pinning dead code.
Delegating collapses both: the factory is now the single construction
path, and the mutation fails 2 tests including the live resolver path.
Also from review:
- maxFreeSockets 32 -> 256 (Node's own default). The lower cap evicted
pooled sockets under concurrency, reintroducing the churn this agent
exists to prevent: at 64 concurrent requests it left 303 sockets in
TIME_WAIT versus 0 at 256.
- Added `timeout` to OPENCODE_AGENT_OPTIONS. Free-socket eviction is
governed by agent.options.timeout, which was unset, so idle sockets
persisted until the peer closed them. `keepAliveMsecs` is the TCP probe
delay, not the idle lifetime.
- resolveProxyTarget() now checks openCodePort before calling
buildOpenCodeUrl instead of relying on it throwing. The port is nulled
on several runtime paths (health-check failure, failed restart), so a
degraded OpenCode made every proxied request pay for a thrown-and-caught
exception — and the getter added a second call per request.
- Test fixtures use :4096 rather than :443; WHATWG URL elides the default
port, so parseInt('') is NaN and env-config rejects that host. The
fixtures modeled a state that cannot reach production.
- The getter-read assertion is now exact (0 at construction, 1, then 2)
rather than >= 2, which would have passed if the getter were read twice
at construction and never per-request.
- listen() rejects on 'error' and servers start inside try/finally, so a
bind failure fails the test instead of hanging to timeout.
A failed nightly run left edits in the maintenance clone, and every later run
correctly refused to work on a dirty worktree, so one failure stalled the whole
pipeline until morning.
Maintenance task commands now recognise a gitignored .maintenance-clone marker.
In a marked disposable clone they discard leftover debris, return to main, and
continue; in a human working copy they still stop and touch nothing.
Add an explicit abort protocol: revert your own edits, confirm the worktree is
clean, release the claim, and report. Restore the honest skip that the
complete-file rule had squeezed out, since a laundered fix is worse than a
documented skip, and describe how to handle a file that is entirely an
external-data boundary instead of inventing generic JSON contracts.
Review follow-up on the #2769 highlight caches.
Fingerprint strength. The block/highlight caches are now global and
content-addressed, so a hash collision no longer mis-colors a block — it returns
a *different* block's rendered HTML and shows the user source they never wrote.
Length + one 32-bit FNV-1a is not enough key space for that failure mode at
session scale. `contentFingerprint` now combines two independent 32-bit
multiplicative hashes with a final avalanche (~64 bits); two multiplies per
character are free next to Shiki tokenization.
Streaming churn. Content addressing made every streaming step of the trailing
`live` block insert a new cache entry, so one long message evicted the settled
`full` blocks the fix exists to keep warm. `full` and `live` blocks now use
separate caches; the live cache is small (32 entries / 2MB) because it only has
to absorb repeat renders of the same step.
Redundant worker-side caches. `markdown-worker.ts` is the only sender to the
Shiki worker, and its client cache is larger than the worker-side ones, so the
worker caches could not serve a hit the client had not already served — they
only duplicated up to 48MB of payloads in a second heap. Removed; the reason
memoization belongs on the client is now documented there, along with why only
`highlightTokens` carries a theme in its key.
Dead `cacheKey` plumbing. `renderMarkdownBlocks` kept a `cacheKey` parameter it
only `void`-ed. Removed it and the now-unused `useMorphdomMarkdown` prop; the
remaining call-site local is renamed `fadeKey` for what it actually keys.
Tests: image-mode cache identity, streaming-does-not-evict-settled-blocks,
live-cache reuse, and a 20k same-length-source fingerprint collision check.
Each new guard was verified to fail without its fix.
Treating every undefined parse as empty config let a file that is not JSON
at all (YAML, plain text) read as {}, so a later write would back it up and
replace it - the same data loss this fix is meant to prevent. Only a
comment-only parse, where ValueExpected is the sole error, counts as empty.
Conflict: packages/ui/src/components/chat/markdown/markdownCore.ts
main added per-image-mode markdown parsers (`imageMode` threaded through
`parseBlock` and into the block cache key); this branch replaced the
identity-keyed block cache with a content-addressed LRU. Resolution keeps the
content-addressed cache and folds `imageMode` into the content key, so the
`inline` and `label` renderings of the same source cannot answer for each other.
* fix(chat): defer composer value writeback during IME composition
The controlled-writeback effect compared the value prop against the
CodeMirror document and, on mismatch, dispatched a wholesale replacement
with the caret forced to the end. While the browser composes (pinyin,
kana, hangul) the uncommitted text lives in the DOM, not in the document,
so the mismatch is expected and the dispatch interrupted the IME session
and jumped the cursor. Skip the writeback while the view is composing,
using CodeMirror's public compositionStarted getter; the composition
commits through its own pipeline and reports via onChange.
Fixes#2527
* fix(chat): preserve external composer writes during IME
* fix(chat): restore composition-wide writeback guard
---------
Co-authored-by: Bohdan Triapitsyn <artmore@protonmail.com>
Maintenance task commands now fill .github/PULL_REQUEST_TEMPLATE.md section by
section instead of inventing their own headings, and follow-up tasks keep the
description true for the final HEAD while preserving hand-added content.
Raise the anti-slop batch window to 60-120 findings and require each selected
file to be finished: remaining findings need an individual specific reason,
shared root causes count once, and difficulty alone no longer justifies a skip.
A half-fixed file otherwise returns as a second pull request over the same code.
Add the maintenance-review command, which reviews every open anti-slop and
react-doctor pull request and fixes the findings directly rather than
commenting, without merging or approving.
Vendor the anti-slop Oxlint plugin at tools/oxlint/anti-slop and register it
in oxlint.config.ts, with Oxlint's own rule categories disabled so ESLint
stays the general-purpose linter.
Add scripts/anti-slop.mjs (bun run deslop) mirroring the React Doctor batch
interface: next-batch, check-batch, active, release, top, file. Batch handoff
directories now double as file claims shared across clones via
~/.openchamber/maintenance-claims, so concurrent maintenance batches from
either pipeline never select the same file.
Harden both scheduled maintenance flows: stop on a dirty worktree, stop on
NO BATCH AVAILABLE, validate per package instead of workspace-wide, and pin
react-doctor to 0.9.12. The anti-slop task command documents concrete
good and bad fixes and forbids laundering types to satisfy a rule.
The token breakdown of an assistant message accumulates across every API
round-trip inside the turn: each tool call re-reads the whole cached
prompt, so input/cache.read add up to several times the context window.
Every context-usage surface summed those fields, which is why the meter
could read 330% of a 1M window whose real fill was 232,872 tokens
(23.3%), and why reopening an older session jumps the readout (#2562).
The server reports the final round-trip's window as tokens.total
(optional in the message schema; opencode 1.18.18 returns it, verified
against its live /session/:id/message API). Prefer it everywhere the
window fill is displayed and fall back to summing only when the server
did not send it: contextTokensFromBreakdown in tokenUtils now owns that
rule, and the context store extractor, sync store getter, work status
panel, context sidebar, VS Code layout, mini chat, and mobile metadata
all use it instead of their own inline sums.
Fixes#2562
Branch status resolves an open PR across the whole fork network first, so a
merged fork PR can never hide an open upstream PR for the same head. Only when
no target has an open PR does the branch's newest closed/merged PR come back,
as history.
The panel shows that history as a compact note and offers creating the next PR
below it, instead of either sticking on a terminal PR or going blank after a
merge. Terminal associations stay persisted for reload continuity but are never
treated as authority: they revalidate on the discovery cadence and on focus.
History is looked up only for the branch's own remote and name, and remembered
per repo+branch, so the extra lookup cannot exhaust the route's resolve budget.
The checks summary and merge-permission lookup are skipped for a closed or
merged PR, where neither is actionable.
Auto-derived project labels were title-cased, turning .ssh into .Ssh and
opencode-claude into Opencode Claude. Show the folder name verbatim in the
sidebar, window title, settings selector and notification templates, and
migrate persisted legacy labels back to the folder name (manual renames
are preserved).
Create no longer aborts when recoverStaleDraftDirectory rewrites the
implicit new-chat draft to the active project during the create probe.
Also rank the Unreleased Chat bullet below the changelog highlights.
Co-authored-by: serkraser <serkraser@gmail.com>
Plugin listing still called readConfigFile per layer, so one unparseable
project file made GET /api/config/plugins and VS Code listPluginEntries
fail. Reuse readConfigLayer isolation and pin comment-only empty parse
in the VS Code suite.
Co-authored-by: serkraser <serkraser@gmail.com>
Keep both Unreleased bullets: the config-wipe Settings fix and the
stale merged-PR Git panel fix from main.
Co-authored-by: serkraser <serkraser@gmail.com>
Address review findings on the #2923 fail-closed parse fix:
- Comment-only files produce ValueExpected with no JSON value; treat that
as empty config instead of INVALID_JSONC. Partial object trees still throw.
- readConfigLayers no longer lets one unparseable layer abort valid sibling
layers. Mutations still fail closed on the custom/user write target.
Co-authored-by: serkraser <serkraser@gmail.com>
Keep both Unreleased changelog bullets: the lastDirectory new-chat
fallback and the Git PR panel stale-status fix from main.
Co-authored-by: serkraser <serkraser@gmail.com>
When a regular new-chat draft inherits a deleted lastDirectory, update
the visible draft target to the active project immediately. lastDirectory
still stays unchanged until session creation succeeds.
Co-authored-by: serkraser <serkraser@gmail.com>
Add store persist/hydrate regressions for terminal branch associations
and a server test that a complete empty open list does not query closed
PRs. Inline the open-only matcher state so the next repo target can win.
Co-authored-by: serkraser <serkraser@gmail.com>
Branch PR status now resolves open PRs only, revalidates closed/merged
associations on a discovery cadence, and clears authoritative empty
results so the panel can self-heal without a manual refresh.
Co-authored-by: Serhii Dziupin <makeittech@users.noreply.github.com>
Bohdan Triapitsyn