Repository navigation
chore: reconcile upstream through fdc7754a - #359
Merged
Merged
Conversation
…enry#2094) (colbymchenry#2280) * fix(windows): hide the git calls that flash console windows (colbymchenry#2094) The two git calls added for exclude-standard scope parity (colbymchenry#1728), `git config --get core.excludesFile` and `git ls-files -o -i --exclude-standard --directory`, ran without windowsHide. Inside the detached, console-less MCP daemon each one got a fresh visible console, so a window flashed several times on daemon start and again on every auto-sync. With Windows Terminal as the default terminal each flash is a full Windows Terminal window. `git config user.name` in the viewer's trails API had the same gap. Add windowsHide: true to all three, and replace the per-file regex guards with an AST guard over every child_process spawn in src/: options passed through a variable or a function parameter are traced back to the object literal that defines them, through every call site. Supersedes colbymchenry#2095 (same fix, rebased on current main). Closes colbymchenry#2094, closes colbymchenry#2096. (cherry picked from commit dd2db41) * docs(changelog): credit the Windows console-window fix Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit cf93d497fd1b38fd9214945eeb55b6b0be9e2007) --------- Co-authored-by: Suharaz <minesuhara@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
colbymchenry#2184) (colbymchenry#2281) * fix(prompt-hook): skip Claude Code <agent-message> subagent hand-backs Claude Code delivers a subagent's report to the parent session as a UserPromptSubmit prompt wrapped in <agent-message from="…">…</agent-message>. The prompt hook treated it as user input and ran a full codegraph_explore on the whole report, which is slow on a multi-KB prose query (past the 30 s hook timeout in the report) and injects context unrelated to the user's question. Add isAgentMessage next to isTaskNotification (colbymchenry#1832) with the same rule: only a prompt that is entirely the envelope is skipped, so a user question that mentions the tag still goes through the gate. Fixes colbymchenry#2184 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 240022d) * docs(changelog): credit the prompt-hook hand-back fix Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: danusha2345 <ewidusoc498@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…enry#2084) (colbymchenry#2282) * fix(mcp): answer server/discover locally in the stdio proxy (colbymchenry#2084) Clients that probe `server/discover` before the legacy `initialize` (Antigravity 2.5) wait for its answer. The proxy forwarded it to the shared daemon, so no reply was written until the daemon connected: a client that closed stdin first got nothing, and with the daemon unavailable the reply came ~6s later as -32603 "daemon unavailable". Answer it in the proxy with -32601 Method not found, which is what sends such a client on to `initialize`. The direct and daemon sessions already answer unknown methods with -32601; the new test pins that too. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 77db6c6) * docs(changelog): credit the server/discover fix Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: danusha2345 <ewidusoc498@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…mchenry#2087) (colbymchenry#2283) * fix(mcp): release an idle explicit project and its writer lock (colbymchenry#2087) A daemon that answered a `projectPath` query for another indexed project opened it in-process and, with no owner there, took its writer lock. The project cache only evicted over its LRU bound or on shutdown, so the lock stayed held for the daemon's whole life: the project's own daemon could not become its writer and `codegraph index` there was refused. Release a cached explicit project once it has gone unused for 10 minutes (CODEGRAPH_PROJECT_IDLE_TIMEOUT_MS; 0 never releases), through the same trim path as LRU eviction, so an active call or a running catch-up still defers it. The next call reopens the project and catches it up (colbymchenry#1835). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit 092d49d) * docs(changelog): credit the idle project release Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: danusha2345 <ewidusoc498@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
…#2088) (colbymchenry#2284) * fix(mcp): return status and stale refusals as text only (colbymchenry#2088) Claude Code shows the model a tool result's structuredContent in place of its content text, under every protocol version. codegraph_status and explore's stale-index refusal both set structuredContent.freshness, so the model saw a JSON object instead of the refusal, its retry guidance or the status report. The text already carries every field, and nothing in the codebase reads structuredContent, so both results drop it. The tests now assert that the field is absent and that the text keeps the refusal line and the stale file list. (cherry picked from commit 18d65f8) * fix(mcp): list every unverified file in the stale refusal (colbymchenry#2088) (cherry picked from commit f1b0ec7) * fix(mcp): keep the stale refusal's unchecked-file list bounded The text-only refusal now names the first 20 files it could not verify and counts the rest, as it did before; an impact answer over a hub symbol can leave thousands unchecked, and every one of them would land in the agent's context. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit c4513a66c9ae6bb10ede80d102d4122ed7f0508d) * docs(changelog): one entry for the text-only results fix Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Aaron Queen <bompus@users.noreply.github.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Brings in upstream colbymchenry#2280-colbymchenry#2284: hidden git calls on Windows, the prompt hook skipping subagent hand-backs, the local server/discover answer, the idle release of a projectPath project, and text-only status results. The fork keeps its own versions where it already had one: - Stale answers still warn and answer, naming every file the freshness check could not verify, instead of refusing. - isHostNotification already covered task notifications; only isAgentMessage is added. Upstream's idle release replaces the fork's interval sweep, so the setting is now CODEGRAPH_PROJECT_IDLE_TIMEOUT_MS. The new windowsHide guard test found five fork-only spawn sites, which now set it. Upstream's idle timer kept whichever deadline was armed first. An older project whose catch-up finished after a newer project armed the timer then waited for the newer deadline. Every trim now re-arms it.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Reconcile upstream through
fdc7754a(colbymchenry#2280–colbymchenry#2284). This is the first of three batches toward9d4e805b. It lands as a merge commit so upstream stays an ancestor.What comes in from upstream:
server/discoveris answered locally (fix(mcp): answer server/discover locally in the stdio proxy (#2084) colbymchenry/codegraph#2282).projectPathproject is released when idle (fix(mcp): release an idle explicit project and its writer lock (#2087) colbymchenry/codegraph#2283).Where the fork keeps its own behavior:
isHostNotificationalready covered task notifications. Only upstream'sisAgentMessageis added.Where the fork changes to match upstream:
CODEGRAPH_PROJECT_IDLE_TIMEOUT_MS(README and site doc updated). The fork's duplicate test and CHANGELOG entry are removed.Fix to upstream's idle release: upstream's timer kept whichever deadline was armed first. Suppose an older project's catch-up was still running when a newer project armed the timer. When that catch-up finished, the older project waited for the newer deadline, up to almost twice the timeout, and held its writer lock that long. Every trim now re-arms the timer. A new fake-timer test (
mcp-idle-release-order.test.ts) fails without the fix: nothing has been released at 3.6 s. With the fix it passes.Changes for the fork:
mcp-server-discover.test.tsno longer passes the WASM runtime flags, since the fork has no WASM parser.windowsHideguard test flagged five spawn sites that only exist in the fork. They now setwindowsHide.fork()passes its options on tospawn(), butForkOptionsdoesn't declare the key, so it is spread in.structuredContententries are folded into the fork's existing ones.Validation: tsc clean. Targeted tests pass, 94/94, across frontload-hook, mcp-server-discover, child-process-windows-hide, mcp-stale-refusal, mcp-status-freshness, mcp-projectpath-lifecycle and readme-sync. The full suite passes: 7,087 tests across 562 files, with 34 existing skips and no failures. Kernel sources are unchanged, so the golden dumps are not affected. An independent review passed on round 3. Round 1 found the idle-timer bug; round 2 asked for a deterministic test in place of the timing-based one.
README rows checked: the upstream merge point (now
fdc7754a), the "About this fork" tables (no row covers these features), and the measured-results rows (unaffected).Follow-up, not in this PR: upstream's
windowsHideguard test accepts a spread that later sets the key to false, as in{ windowsHide: true, ...{ windowsHide: false } }. No source file does this today.