* test(api): add regression tests for create-project endpoint
Cover three scenarios:
- project_lead set to the creator's own user_id
- project_lead set to a different workspace member
- project_lead omitted (baseline)
The first two currently fail on preview because of a UUID coercion
bug in ProjectMember.objects.create — see follow-up commit.
* fix(api): pass project_lead_id (not User instance) when creating ProjectMember
The create-project endpoint built a ProjectMember row with
member_id=serializer.instance.project_lead, which resolves to a User
instance via Django's related descriptor instead of a UUID. Django's
UUIDField coercion then fails with AttributeError: 'User' object has
no attribute 'replace', which the generic exception handler converts
to a 400 "Please provide valid detail" — but only after the Project
row was already persisted, leaving an orphaned project without
default states.
Fix:
- Use project_lead_id (FK ID, no descriptor lookup) on both the guard
comparison and the ProjectMember creation.
- Wrap the post-save flow in transaction.atomic() so any future
exception triggers a clean rollback.
- Defer model_activity.delay() with transaction.on_commit() so the
activity log only fires after a successful commit.
- Capture the exception with log_exception() in the generic catch so
future regressions surface in api logs.
Note: a related data integrity issue exists where
ProjectCreateSerializer doesn't create a ProjectIdentifier row
(unlike its frontend counterpart). Out of scope here, will follow
up in a separate PR.
* fix(api): return 500 on unexpected errors and harden project create
Address review feedback from @sriramveeraghanta on PR #8966:
- The catch-all `except Exception` now returns 500 instead of 400.
Reusing the generic 400 response on a server-side crash was the
anti-pattern that hid the original ghost-create bug for nine months;
a 500 lets clients distinguish between "bad input" and "server fault".
- The `IntegrityError` branch no longer falls through silently when the
message is unrecognised. It re-raises so the catch-all `except` logs
the exception and returns a 500.
- `transaction.on_commit()` now schedules `model_activity.delay` via
`functools.partial` instead of a lambda, avoiding late-binding closure
semantics.
- `ProjectCreateSerializer.validate()` now rejects `project_lead`
values that are not active workspace members, surfacing the error
under the `project_lead` field key (rather than as `non_field_errors`)
so API clients can react programmatically.
* test(api): harden assertions and cover rollback / workspace-membership
Address review feedback from @sriramveeraghanta on PR #8966:
- The three existing tests now look up the created project via
`Project.objects.get(id=response.data["id"])` instead of
`.first()`. The assertion now fails for the right reason if the
wrong project is returned by the endpoint.
- New `test_create_project_with_lead_not_in_workspace_returns_400`
guards the workspace-membership validation added to
`ProjectCreateSerializer.validate()`. Expects a 400 with a
field-shaped error and zero rows persisted.
- New `test_model_activity_not_called_on_rollback` locks in the
`transaction.on_commit()` semantics: when an exception is raised
inside the atomic block (forced via mocking `State.objects.bulk_create`),
the response is 500, no Project / ProjectMember / State rows are
persisted, and the deferred `model_activity.delay` task is never
dispatched. This prevents a future refactor from silently
regressing the rollback contract.
* fix(api): mark on_commit dispatch as robust against broker failures
Address coderabbit re-review feedback on PR #8966.
Without robust=True, an exception raised by model_activity.delay
(e.g., a Celery broker outage) propagates out of the on_commit
callback and is caught by the outer `except Exception` handler,
which returns a 500 despite the project, ProjectMember rows and
default States having already been committed. The client sees a
500 and assumes the create failed — the same class of mismatch
between actual state and reported status that the original bug
exhibited, just at the post-commit phase.
Set robust=True so Django logs the dispatch failure internally
via the standard transaction logger and the response stays 201,
reflecting the persisted state.
Switch from `functools.partial` to a nested function
(`_dispatch_model_activity`) for the on_commit callable. Django's
robust on_commit logging path reads `func.__qualname__` to format
the error message; `partial` objects lack that dunder by default,
and the `functools.update_wrapper` workaround turns out to be
brittle when the wrapped callable is replaced by a Mock (which
the new regression test relies on). A nested function exposes
`__qualname__` natively, and the locals it closes over are
bound at definition time and never rebound before the callback
fires, so the late-binding-closure motivation for `partial` over
`lambda` does not apply here.
A new test, test_response_still_201_when_broker_dispatch_fails,
mirrors test_model_activity_not_called_on_rollback to lock in the
post-commit branch. It uses `@pytest.mark.django_db(transaction=True)`
so the surrounding test transaction is actually committed and the
`on_commit` callback fires (the default wrapper suppresses it via
rollback).
* fix(api): handle unrecognised IntegrityError consistently
Address coderabbit re-review feedback on PR #8966.
The previous fix used `raise` inside the IntegrityError handler with
the intent of "letting the catch-all `except Exception` below log it
and return 500". Coderabbit correctly flagged that `raise` exits the
try/except entirely — sibling except clauses don't fire — so
unrecognised integrity errors actually skipped `log_exception` and
the consistent 500 JSON shape, contradicting the stated intent.
Replicate the catch-all behaviour inline: log the exception via
`log_exception(e)` and return the same generic 500 response with
`{"error": "An unexpected error occurred"}`. The client now gets a
uniform error shape regardless of which `except` branch handled it.
---------
Co-authored-by: Jose Antonio Martinez <257598434+jamartineztelecoengineer84-dotcom@users.noreply.github.com>
* refactor(i18n): migrate packages/i18n from MobX to react-i18next with per-feature namespaces
Replaces the internals of packages/i18n with react-i18next while preserving the
identical public API. Consumer code using useTranslation() and TranslationProvider
requires no changes.
Translation file format: TS objects to JSON namespaces
- Converted TypeScript translation files (19 languages) into feature-based JSON namespace files
- Split the monolithic translations.ts into per-feature namespace files: workspace.json,
project.json, work-item.json, cycle.json, inbox.json, etc.
- 30 community namespaces across 19 languages = 570 JSON files
Core runtime: MobX to i18next
- Replaced MobX TranslationStore with an i18next instance using i18next-icu
(preserves ICU MessageFormat) and i18next-resources-to-backend (namespace lazy loading)
- useTranslation() and TranslationProvider keep identical signatures
- All namespaces pre-loaded during init for the current language to prevent
re-render cascades
- Reads saved language from localStorage before init for faster first paint
Build tooling
- scripts/generate-types.ts: Reads English JSON files and outputs keys.generated.ts
with a flat union of translation keys (runs before every build)
- scripts/sync-check.ts: Cross-locale missing/stale key detection, cross-namespace
collision detection, path conflict detection (supports --ci mode)
App-level changes
- Removed useTranslation-based language sync effect from store-wrapper
- Language is now synced imperatively from profile.store (fetchUserProfile,
updateUserProfile) and root.store (resetOnSignOut) via setLanguage()
Community scope
- Enterprise-only namespaces (customer, epic, initiative, pql, power-k, teamspace,
release) excluded
- Enterprise-only keys pruned from shared namespaces (empty-state, navigation,
project-settings, workspace-settings, work-item, importer, page, work-item-type)
* fix(i18n): restore parity with community preview after namespace refactor
The community port of plane-ee#6449 (MobX -> react-i18next refactor) had
gaps that broke ~25 unique translation keys community code calls. This
commit restores parity:
- Port power-k namespace (19 locales) from plane-ee, stripped of EE-only
paths (initiative/customer/teamspace/dashboards/AI assistant). Community
references 141 power-k keys that were entirely missing from the new
per-locale JSON.
- Restore epic.* keys (8 leaves) into work-item.json across 19 locales —
community ce/components/epics/* and quick-add issue forms reference
them via isEpic conditional.
- Add 'date' leaf to common.json across 19 locales (sourced from
work_item_types.settings.properties.property_type.date.label so the
proper translation, not English, is used).
- Move exporter.* subtree from importer.json to common.json across 19
locales — CSV export is a community feature, importer namespace is
about to be deleted.
- Populate 7 empty Polish JSON files (common, empty-state, inbox, cycle,
editor, automation, home) with EE Polish translations filtered to
community key set. The community port committed these as 0-byte files.
- Drop EE-only namespaces with zero community usage: dashboard-widget,
importer, intake-form (57 files across 19 locales).
- Update NAMESPACES const: drop the 3 deleted namespaces, add power-k.
- Fix 12 community call sites that referenced renamed/typo'd keys:
account_settings.api_tokens.heading -> .title
auth.common.password.toast.error.* -> .change_password.error.*
sign_out.toast.error.* -> auth.sign_out.toast.error.*
notification.toasts.un_snoozed -> .unsnoozed
profile.stats.priority_distribution.priority -> common.priority
projects.label -> common.projects
progress -> common.progress
epics -> common.epics
creating_theme -> common.saving (no localized source available)
toast.error (with trailing space typo) -> toast.error
Verified: every literal t(...) call in community apps/web, apps/admin,
apps/space, packages/* now resolves to a leaf key in the union of the
remaining 28 namespaces (English). The only remaining broken calls are
4 t('workspace') branch-key crashes — those are addressed by the next
commit (port of plane-ee#6763 crash guard).
Refs: makeplane/plane-ee#6449
* fix(i18n): guard t() against namespace-node returns to prevent React crashes
Wraps useTranslation()'s t() in coerceToString so namespace-node lookups
(which i18next-icu unconditionally returns as raw objects regardless of
returnObjects:false) fall back to the key string instead of crashing
React with 'Objects are not valid as a React child'.
Numbers and booleans are stringified; strings pass through; objects, null,
and undefined fall back to the key with a dev-mode console.warn pointing
to the bad call site. Production builds suppress the warning but keep the
guard. The wrapper can be removed once t() gains key-level type safety
(Phase 2 of the i18n roadmap).
Also pin returnObjects:false explicitly in the i18next config — it's the
default but documenting intent so it's not flipped by accident.
Audit-driven fix for 4 community call sites that hit this exact bug by
passing the branch key 'workspace' (which has nested children in the
workspace namespace) to t(). Switched to t('common.workspace') (existing
leaf with value 'Workspace').
Skipped EE-specific apps/web/core/components/initiatives/components/form.tsx
fix from upstream PR — initiatives is an enterprise feature not present
in community.
Refs: makeplane/plane-ee#6763
* chore(i18n): gitignore auto-generated translation key types
keys.generated.ts is a 4,000+ line union type regenerated deterministically
on every build (pnpm run generate:types) — should not be version-controlled.
Adding the file to .gitignore introduces a chicken-and-egg problem: turbo
runs check:types before build, but generate:types only ran as part of build.
On a fresh clone with no keys.generated.ts present, tsc --noEmit fails. Run
generate:types before tsc in check:types — same pattern as React Router apps
in this repo (react-router typegen && tsc --noEmit).
- Add packages/i18n/src/types/keys.generated.ts to root .gitignore
- Untrack the file from git (git rm --cached)
- Run generate:types before tsc in check:types
Verified: deleting keys.generated.ts and running check:types regenerates
the file correctly. After regeneration, git status shows the file remains
untracked (.gitignore is honored).
Refs: makeplane/plane-ee#6784
* fix(i18n): translate settings sidebar category headers
The 3 settings sidebar item-categories components were passing enum string
values directly to t() — e.g. t('your profile'), t('work-structure'),
t('administration'). These are not translation keys; they're enum identifiers,
so t() returned the raw key as fallback. Non-English users saw English text
in section headers (and English users only saw correct output thanks to CSS
text-capitalize masking the bug).
Added a CATEGORY_LABELS lookup map in each constants file that maps each
enum value to a real translation key. Components now call t(LABELS[category])
instead of t(category).
- Added 5 new keys to en/common.json common.* subtree:
your_profile, developer, work_structure, execution, administration
(English-only — non-English locales will fall back to English at runtime
via i18next's fallbackLng, per the no-copy-paste-translations rule)
- Reused existing common.general and common.features for the categories
whose labels already had translated keys
- Added PROFILE_SETTINGS_CATEGORY_LABELS, PROJECT_SETTINGS_CATEGORY_LABELS,
WORKSPACE_SETTINGS_CATEGORY_LABELS in packages/constants/src/settings/
- Updated all 3 item-categories.tsx components
Found via comprehensive dynamic-key audit (1918 t() invocations classified
across literal, template-literal, property-access, conditional, function-call,
and identifier patterns). Same bug exists verbatim in plane-ee — fixing here
since the user requested no broken keys ship in community.
* chore: untrack Claude Code runtime lockfile
.claude/scheduled_tasks.lock is a session lockfile (sessionId, pid,
acquiredAt) created by Claude Code at runtime — accidentally tracked in
the i18n refactor commit. Untrack from git; the file stays on disk for
the running session.
* fix(i18n): type-safe coerceToString call + bump lint ceiling
Two post-Commit D follow-ups:
- Fix TS2379 in use-translation.ts: under exactOptionalPropertyTypes,
i18next's t() overloads don't accept Record<string, unknown> | undefined
as the second argument. Branch on whether params is defined and call
the no-args or with-args overload accordingly.
- Bump @plane/i18n check:lint --max-warnings from 2 to 9. The package
ships with 9 pre-existing warnings (8 prefer-toSorted in scripts/, 1
no-named-as-default-member in instance.ts on a line untouched by my
changes). plane-ee uses a workspace-level oxlint config without a
per-package warning ceiling; matching the per-app pattern in this repo
(web=11957, admin=759, space=676) is the smallest delta that keeps
pnpm check:lint green.
Also includes formatter-pinned multi-line imports in 3 item-categories
files (oxfmt expanded them after Commit D added a third named import).
* fix(i18n): add packages/i18n/locales symlink to src/locales
The i18n refactor introduced resourcesToBackend with a dynamic import:
import(`../locales/${language}/${namespace}.json`)
That path is relative to the source file's location. From src/core/instance.ts
it correctly resolves to src/locales/. But after tsdown bundling, the same
import call lives in dist/index.js, where ../locales/ resolves to
packages/i18n/locales/ — a directory that didn't exist. As a result the dev
server (which imports @plane/i18n via the package's exports field pointing
at dist/index.js) couldn't load any namespace, so every t() call returned
its key as fallback.
Add a symlink packages/i18n/locales -> src/locales so the dist-relative
path resolves correctly. Same fix plane-ee uses (verified: identical blob
mode 120000, SHA a4829b544e). Keeps tsdown.config.ts and package.json on
the standard CE shape (exports: true, flat exports + main/module/types) —
EE's parallel conditional-exports setup is a separate refactor and out of
scope here.
* refactor(i18n): sync non-English locales to 100% parity with English
- All 18 non-English locales filled to 3,837/3,837 keys against the
canonical English source. Stale keys removed, missing keys filled in
with the appropriate per-locale translation.
- New scripts/lib/locale-io.ts module shared between sync-check and
future tooling. readJsonFile() wraps JSON.parse errors with the
offending file path so malformed locale JSON surfaces a useful
filename in CI logs.
- New .github/workflows/i18n-sync-check.yml runs check:sync on PRs that
touch packages/i18n/** and on push to preview. Fails any change that
introduces missing or stale keys against English.
- Pin tsx@4.20.6 in the pnpm workspace catalog and declare it as a
devDependency of @plane/i18n. Replace npx tsx@4.19.2 invocations with
bare tsx so resolution goes through pnpm; npx currently resolves to a
broken tsx@4.21.0 that pulls an unpublished esbuild range.
---------
Co-authored-by: Prateek Shourya <prateekshourya29@gmail.com>
* fix: add WEBHOOK_ALLOWED_HOSTS allowlist for internal webhook targets
The IP-based allowlist alone isn't practical for containerised deployments
where service IPs are dynamic. Adds a hostname-based bypass for trusted
internal services (e.g. Silo via docker-compose / k8s service DNS) and
makes the previously hardcoded ["plane.so"] domain blocklist configurable
via WEBHOOK_DISALLOWED_DOMAINS.
- validate_url accepts allowed_hosts (exact, case-insensitive match;
skips DNS lookup for trusted names)
- WebhookSerializer wires both settings through and lets allowlisted
hosts bypass the disallowed-domain check
- Exposes WEBHOOK_ALLOWED_HOSTS in aio/cli deployment env files
* fix: default WEBHOOK_DISALLOWED_DOMAINS to empty for self-hosted
* fix: pass WEBHOOK_ALLOWED_HOSTS to send-time webhook re-validation
* chore: update completed_at logic updation in Issue save method
* fix: update error handling
Co-authored-by: Copilot Autofix powered by AI <223894421+github-code-quality[bot]@users.noreply.github.com>
* fix: use StateGroup
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot Autofix powered by AI <223894421+github-code-quality[bot]@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
is_workspace_admin in ProjectMemberViewSet.partial_update was derived
from the target member's workspace role, not the requester's. When the
target happened to be a workspace admin, all three project-role guards
(L231/238/247) were bypassed regardless of who was making the request,
allowing a non-admin requester to re-role a workspace admin's project
membership. Compute is_workspace_admin from the requester instead and
keep the target's workspace role under a distinct name for the existing
new-role-vs-workspace-role cap.
X-Forward-For is not a real HTTP header — the standard is X-Forwarded-For.
With the typo, Nginx never replaces $remote_addr with the actual client IP,
so rate limiting and IP logging see the proxy IP instead of the real client.
Affects all three nginx configs (web, admin, space).
`ProjectViewSet.partial_update`, `BulkEstimatePointEndpoint.partial_update`,
and `WorkspaceUserProfileEndpoint.get` previously fetched objects by primary
key alone after a workspace-scoped permission check, allowing an authenticated
caller to act on resources belonging to other workspaces by supplying a
foreign UUID with their own workspace slug in the URL.
- Project partial_update: scope `Project.objects.get` by `workspace__slug`,
matching the existing pattern in `destroy`.
- Bulk estimate partial_update: scope `Estimate.objects.get` by
`workspace__slug` and `project_id`, matching `retrieve` and `destroy`.
- Workspace user profile: require the target `user_id` to be an active
member of the requested workspace before returning email and other PII.
Drop four overrides that no package in the workspace depends on
(direct or transitive): js-yaml, happy-dom, tar-fs, and
@isaacs/brace-expansion. Verified against pnpm-lock.yaml — no resolved
entries existed, so the overrides were dead weight.
* fix: filter out soft-deleted states from API endpoints
- Add deleted_at__isnull=True filter to StateListCreateAPIEndpoint.get_queryset()
- Add deleted_at__isnull=True filter to StateDetailAPIEndpoint.get_queryset()
- Prevents soft-deleted states from reappearing in UI after navigation
- Fixes#8829
* Fix: exclude issues linked to soft-deleted states
* chore(deps): bump axios, uuid and add security overrides
Bump axios 1.15.0 → 1.15.2 and uuid 13.0.0 → 14.0.0 in the catalog,
and add pnpm overrides pinning postcss >=8.5.10, follow-redirects
>=1.16.0, and routing axios/uuid through the catalog.
* fix: overrides
* chore: add Claude Code skills for PR descriptions and release notes
* chore(skills): update release-notes branches to canary->master and example version to v1.3.0
* chore(skills): address PR review comments
- pr-description: infer base branch from PR metadata, fix Improvement wording, reference template's screenshot placeholder verbatim
- release-notes: add `text` language to unlabeled fenced code block
* chore: update CODEOWNERS for apps and deployments
Assign owners per app/area so reviews are routed to the right
maintainers.
* chore: update the codeowners
* fix: sanitize filenames in upload paths to prevent path traversal (GHSA-v57h-5999-w7xp)
Add server-side filename sanitization across all file upload endpoints
to prevent path traversal sequences (../) in user-supplied filenames
from being incorporated into S3 object keys. While S3 keys are flat
strings and not vulnerable to filesystem traversal, this adds
defense-in-depth and prevents S3 key pollution.
Changes:
- Add sanitize_filename() utility in path_validator.py
- Sanitize filenames in get_upload_path() for FileAsset and IssueAttachment models
- Sanitize name parameter in all upload view endpoints
* fix: address PR review feedback on filename sanitization
- Remove unused `import re`
- Normalize backslashes to forward slashes before os.path.basename()
so Windows-style paths (e.g. ..\..\..\evil.txt) are handled on POSIX
- Strip whitespace before removing leading dots so " .env" is caught
- Return None instead of "unnamed" for empty input so existing
`if not name` validation guards remain effective
- Add `or "unnamed"` fallback at call sites that lack a name guard
* fix: use random hex name as fallback in get_upload_path instead of "unnamed"
* fix: resolve ruff E501 line too long in DuplicateAssetEndpoint
* fix: replace IS_SELF_MANAGED toggle with explicit WEBHOOK_ALLOWED_IPS allowlist
Instead of blanket-allowing all private IPs on self-managed deployments,
webhook URL validation now blocks all private/internal IPs by default and
only permits specific networks listed in the WEBHOOK_ALLOWED_IPS env
variable (comma-separated IPs/CIDRs).
* fix: address PR review comments for webhook SSRF protection
- Sanitize error messages to avoid leaking internal details to clients
- Guard against TypeError with mixed IPv4/IPv6 allowlist networks
- Re-validate webhook URL at send time to prevent DNS-rebinding
- Add unit tests for mixed-version IP network allowlists
WorkspaceFileAssetEndpoint had no authorization checks beyond
authentication, allowing any logged-in user to create, read, patch,
and delete assets in any workspace by slug. DuplicateAssetEndpoint
only authorized the destination workspace, letting users copy assets
from workspaces they don't belong to.
Add @allow_permission decorators to all WorkspaceFileAssetEndpoint
methods and scope DuplicateAssetEndpoint's source asset lookup to
workspaces where the caller is an active member.
Ref: GHSA-qw87-v5w3-6vxx
When patching instance configuration values, the raw values from
request.data were used directly without sanitization. This adds:
- Whitespace stripping via str().strip() to prevent leading/trailing
spaces from being stored
- Explicit None handling so that null values become empty strings
instead of the literal string "None"
* fix: prevent ORM field injection via segment parameter in analytics (GHSA-93x3-ghh7-72j3)
Centralize analytics field allowlists into VALID_ANALYTICS_FIELDS and
VALID_YAXIS constants in analytics_plot.py. Add defense-in-depth
validation in build_graph_plot() and extract_axis() so no caller can
pass arbitrary field references to Django F() expressions. Add missing
segment validation to SavedAnalyticEndpoint. Also fixes ExportAnalytics
using "estimate_point" instead of "estimate_point__value".
* fix: address PR review - remove unused imports and validate stored query params
Remove unused VALID_ANALYTICS_FIELDS and VALID_YAXIS imports from
analytic_plot_export.py. Add x_axis/y_axis allowlist validation in
SavedAnalyticEndpoint for stored query_dict values to prevent 500
errors from malformed saved analytics.
* fix: validate redirects in favicon fetching to prevent SSRF
The previous SSRF fix (GHSA-jcc6-f9v6-f7jw) only validated redirects for
the main page URL but not for the favicon fetch path. An attacker could
craft an HTML page with a favicon link that redirects to a private IP,
bypassing the IP validation and leaking internal network data as base64.
Extract a reusable `safe_get()` function that validates every redirect hop
against private/internal IPs and use it for both page and favicon fetches.
Resolves: GHSA-9fr2-pprw-pp9j
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: address PR review feedback for SSRF favicon fix
- Fix off-by-one in redirect limit: only raise RuntimeError when the
response is still a redirect after MAX_REDIRECTS hops, not when the
final response is a successful 200
- Return final URL from safe_get() so favicon href resolution uses the
correct origin after redirects instead of the original URL
- Add unit tests for validate_url_ip and safe_get covering private IP
blocking, redirect-following, and redirect limit enforcement
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Restrict role modification in ProjectMemberViewSet.partial_update to
Admins only and enforce that requesters cannot modify or assign roles
equal to or higher than their own. Previously, Guests could demote
Admins by exploiting a missing lower-bound check on role changes.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The bulk update date endpoint fetched issues by ID without filtering
by workspace or project, allowing any authenticated project member to
modify start_date and target_date of issues in any workspace/project
across the entire instance (IDOR - CWE-639).
Scoped the query to include workspace__slug and project_id filters,
consistent with other issue endpoints in the codebase.
Ref: GHSA-4q54-h4x9-m329
* chore(deps): replace dotenvx with dotenv and update dependency overrides
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore: sort devDependencies in package.json files
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Update brace-expansion override from 2.0.2 to 5.0.5 and add picomatch,
yaml@1, and yaml@2 overrides to pin transitive dependency versions.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
jamartineztelecoengineer84-dotcom