Repository navigation
fix: backport 0.12.1 wide batch 6 (11 user-facing fixes + 1 prerequisite) - #32984
Conversation
…#30880) The spawn_sub_agents description said 'Preferred over spawn_run when you need results before continuing', which steered models to pick the blocking tool by default. This caused dashboard chats to block while sub-agents ran (issue #30851). The description now says 'Use spawn_run instead unless the user explicitly asks to wait' -- spawn_run keeps the chat responsive and shows a task card. PR #29284 made KAS the default harness, making this blocking behavior visible to more users. Closes #30851 Co-authored-by: Raymond Chen <chenmingwei23@users.noreply.github.com> (cherry picked from commit 559ae18)
…ngest (#31037) POST /api/knowledge/sources/{id}/ingest-text passes the body's name to the pipeline, which reads it as a file path, so a number, boolean, list or object name failed the whole ingest with a 500. A list or object namespace failed the same way when SQLite could not store it. Fall back to the source's own name and to the default namespace instead. Values that work today are unchanged. Fixes #31032 (cherry picked from commit 8a82724)
…le (#31107) An installed.json or approved-grants.json whose bytes are not UTF-8 raised UnicodeDecodeError out of the app manager's record readers, which catch only bad JSON and OS errors. Such a file now reads like a corrupt one: the install record as not installed, the grants file as approving nothing. Closes #28955 Co-authored-by: Gary Jennings <8486704+dodgeblaster@users.noreply.github.com> (cherry picked from commit 3922b02)
Settings > MCP reports the result of "Pre-install npm servers" with
"{{ready}} server(s) pre-installed." and "Could not pre-install
{{failed}} server(s)." The (s) shortcut reads wrong for one server in
English and is copied into French, Spanish and Portuguese as
serveur(s) and servidor(es), while Russian had to dodge the plural
with a colon.
Make both strings plural catalog keys that take the count, with the
forms each language needs, so one server reads "1 server
pre-installed." and Russian gets 1 сервер, 3 сервера, 5 серверов.
(cherry picked from commit 7785e9c)
Originally authored on main by Rohan Khanderia (RohanK6); authored here with the public identity because the commit guard accepts only that identity on a hand-resolved pick.
* fix: count unresolved npm pre-installs as failures * test: match the pluralized pre-install failure count --------- Co-author on main: Jacob Largent. (cherry picked from commit 4f9738e)
…#30848) A store made read-only by file permissions (chmod a-w), not by a read-only mount, failed kirocrew snapshot with "attempt to write a readonly database" while the same layout on a read-only mount gave rc 0. The source read already handled the read-only case: a closed WAL database that SQLite cannot open mode=ro is read immutable. The failure was on the DESTINATION. The staging tree walk copies each file with the SOURCE file's mode, so the byte copy of a 0o444 store database is itself 0o444, and the read-write open that writes the consistent copy over it is refused. On a read-only mount the files keep an ordinary mode, so that copy stays writable. Restore owner write on the staged copy before the consistent copy overwrites it. The copy is this process's own file in its private staging tree; the live store is never touched. Covered for WAL and delete journal modes. Co-authored-by: kirocrew-agent <agent@kirocrew.local> (cherry picked from commit 9694b24)
…te (#31065) ensure-node.sh records the node bin dir it chose in <data home>/node-bin-dir, but the process that ran it kept its own PATH. On a git install whose Node came from that script (a mise install off the shell's PATH), kirocrew update then looked npm up on PATH, printed "npm not found -- skipping frontend build" and still reported success, leaving the previous dashboard bundle in place. After the Node check, the update now puts env.node_bin_dirs() first on its own PATH, so the build's npm lookup and npm's own node both find that toolchain. The gateway's boot-time Node repair is left as it was. (cherry picked from commit 55e8007)
…1052) On a host that reaches the internet only through HTTPS_PROXY, saving a new Discord, Telegram or Webex bot token reported the service as unreachable and stored the token unverified, while the running channel connected fine. The token checks opened a plain aiohttp session, which ignores the proxy environment, and dialed directly. Each check now passes the proxy its own channel client uses: the Discord and Webex clients' _resolve_proxy, and for Telegram the client's forward-proxy rule, which skips the proxy for a loopback API base so the token never leaves the machine. The messaging spec names the rule for each save route. Part of #20185 (cherry picked from commit 81d35e1)
…0992) An app's Open action decided "remote" from the gateway's own environment only. The installed Linux service is a system unit, so its environment never names a display server, and every Open on a desktop host answered "Kiro Crew is running remotely" instead of launching. When the gateway names no display, ask the desktop session the user's systemd manager names, the same checked lookup the file manager already uses, and launch when it is live. The launcher gets that session's display, X cookie and desktop name; the session never supplies a bus address, so the child's bus reach is unchanged. (cherry picked from commit 0f93202)
…31081) With "Show Thinking Inline" off (the default), a finished turn folds its working steps behind "Worked through N steps", but every folded tool pill stayed mounted: hidden at zero height, still in the DOM and still subscribed to the store. A turn with 100 tool calls kept about 2,000 hidden elements, and the transcript keeps several such turns mounted, so the chat page's DOM grew with every tool-heavy turn. A settled fold now drops the content of its folded tool rows in every fold mode, as the fold already did with "Show Thinking Inline" on. The row element stays, so opening the fold mounts the pills again; rows that can still go live (running, an MCP App, a pill the user opened) keep their content as before. Part of #23281 (cherry picked from commit 323c4f6)
Design Review (Opus 5.5) — ✅ PASSDesign-level review of Design-Verdict: PASS Nothing blocks: these are faithful cherry-picks of fixes already merged on Notes
[DESIGN-REVIEWED] 7b1fda8 Description read[DESCRIPTION-READ] bf6cd9c55029a186266c51de607f1b182f7d7a15ad660278763e985388d35499 That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
UX Review (Opus 5.5) — ✅ PASS (a BLOCK with no Trigger and no Outcome does not block)UX-level review of UX-Verdict: PASS cannot evaluate: missing screenshots for the changed MCP pre-install messages and the tool-row fold. No blind read was done because the PR attaches or commits no image. Evidence gaps
Notes
[UX-REVIEWED] 7b1fda8 Description read[DESCRIPTION-READ] 9d549df1da917c501ecaec1c628baea56d032d3360e143d07a43e0c660fed0f5 That sha256 names everything this verdict read: the PR description, plus the 3 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
First Principles Review (Opus 5.5) — ✅ PASSPremise-level review of First-Principles-Verdict: PASS Nothing blocks: each item is a cherry-pick of a fix already merged on Notes
What this change shipsInventory (10 items) — 9 justifiedIntent: Bring eleven user-facing 0.12.0 fixes, plus one prerequisite, from
[FIRST-PRINCIPLES-REVIEWED] 7b1fda8 Description read[DESCRIPTION-READ] 1933959ea2c6c6891bec4131ea55eebbcc56b79eb8f568d03c10c8d56c40aaf4 That sha256 names the PR description this verdict read. Recompute it with |
Opus 5.5 Review — ✅ no blocking findingsReviewed Verdict parsed from the review's SHA-scoped output markers for commit False positive or not applicable? A repository writer can comment: |
GPT 6.1 Review — ✅ no blocking findingsGPT 6.1 completed its review of This comment is updated in place on each push. Review detailsNo findings. False positive or not applicable? A repository writer can comment: |
Problem / Motivation
Goal: Bring eleven fixes for problems a 0.12.0 user can hit, merged on
mainafterv0.12.0, ontorelease/0.12.0, with the one main commit one of them needs first, one commit each, in a PR of its own.Why it matters
Each fix changes behaviour a 0.12.0 user can run into today. Every pick applied to
release/0.12.0, with only the catalog hunk listed under "What changed" resolved by hand, so the code each one replaces is on release.allowed_wa_idsmakes the agent refuse to DM you as "ambiguous".nameis not a string.memoryis not an object.ensure-node.sh.Not a goal
main; nothing is changed only here.CHANGELOG.mdchange.release/0.12.0in other batches while this one was being built.What changed (motivation → approach → change)
Each main squash commit was cherry-picked with
-xontorelease/0.12.0@19bd79b4a8, in merge order onmain, one commit each. #30714 sits right before #31156: #31156's component line and test callresolve_all_failedwithcount, which only #30714's plural catalog entry reads. Without it, #31156's new test fails on release.559ae18e98bf3b5c407d6c62426388a2cd9554278a8272489f1d385984d23922b02a5eaa5300bb337785e9c4f0eadcd6151a4f9738e6c503b5a186429694b2482ab596a41a92b91ee25926d4f5a9451e55e8007971118bf0917581d35e178fcb904582700f93202938dd72dcad10323c4f65ec7b1fda8da7Hand-resolved hunk, not product code: #30714 in
website/src/i18n/pluralKeys.json. Its two keys (resolve_all_failed,resolve_ready) were appended. ThesecurityPanel.rule_countline beside them on main comes from a commit not on release, so it was left out. #30714 is authored with the public identity because the commit guard refuses the original address on a hand-resolved pick; its message names Rohan Khanderia. #31156's co-author trailer carried an internal host address and is rewritten as a plain "Co-author on main" line.Backwards compatibility
Compatible: each change is the one already merged on
main, and its compatibility note applies unchanged.Removes nothing: no setting, key or caller is removed. #31197 refuses a model change while config
memoryis not an object, which today fails with a raw TypeError instead.Tests
Proof: not provable -- other: a backport of main commits whose proofs ran on their own PRs
release/*, so the related set was run here. Every touched test file plus the test files that import a touched source module, 81 files, gave 3,338 passed and 4 failed. The same 4 fail on a cleanrelease/0.12.0@19bd79b4a8on this host: threetest_cli.py::TestDoctorcases andtest_gateway_secret_senders_prove_peer.py::TestArtifactSendsThroughThePeerProof::test_list_goes_through_the_proof.tsc -p tsconfig.app.json --noEmitpasses. vitest on the touched test files gave 114 passed (mcpResolveCountPlural,McpManagement.resolveOnce, theTurnBlockfiles).gen-ui-index --checkandcheck-i18n-stringspass.mypy src/kiro_crewandcheck_harness_parity.pypass.check_file_line_capfails only ondashboard/chat_api/resume.py, which is already over the cap on release and is not touched here.Manual verification
N/A - each fix was verified on its own main PR. The 0.12.1 RC soak is the end-to-end check.
Screenshots / video
Why no screenshot: the frontend picks (#30714, #31156, #31081) carry the same rendered change as their main PRs, which show it.
Related Issues
Backport to the 0.12 line for 0.12.1 of #30880, #26915, #31037, #31107, #30714, #31156, #30848, #31197, #31065, #31052, #30992 and #31081. Source issues: #30851, #26912, #31032, #28955, #30706, #30351, #30397, #31192, #29815, #30068.
Sibling sweep
Siblings: none -- each source PR carried its own sibling sweep; this PR only carries those fixes.
Pattern harvest
Not generalizable: this PR only carries fixes already on
main. The source PRs have their own harvest.Checklist
feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)Contribution License Agreement
Contributions are accepted under the project's Apache-2.0 license; see the Licensing section of CONTRIBUTING.md.