feat(slack): add channelTypes filter for DM-specific spawners - #102
Draft
anomalogravity[bot] wants to merge 65 commits into
Draft
feat(slack): add channelTypes filter for DM-specific spawners#102anomalogravity[bot] wants to merge 65 commits into
anomalogravity[bot] wants to merge 65 commits into
Conversation
* feat: add github and linear webhook events * fix: build images * Fixed Issues 1. P1 - MaxConcurrencyError HTTP Response ✅ Fixed ServeHTTP to properly handle MaxConcurrencyError with HTTP 503 and Retry-After header Previously returned generic HTTP 500, now correctly returns 503 for rate limiting 2. P1 - Empty APIVersion/Kind in Owner References ✅ Fixed owner reference creation to use client.Scheme().ObjectKinds() to get proper GVK Added missing Controller and BlockOwnerDeletion fields for proper garbage collection Previously had empty strings, now has correct API version and kind 3. P1 - TOCTOU Race Condition in Idempotency ✅ Replaced separate IsProcessed()/MarkProcessed() with atomic CheckAndMark() Eliminated race window where two requests with same delivery ID could both pass the check Now uses single lock operation for check-and-set 4. P2 - Invalid Kubernetes Names from Truncation ✅ Fixed task name generation to handle nil/empty ID values safely Added strings.TrimRight(taskName[:63], "-.") to ensure names don't end with invalid characters Prevents server-side validation failures 5. P2 - BodyContains Filter Ignored for IssuesEvent ✅ Fixed GitHub filter to check issue body for BodyContains on IssuesEvent Previously only checked comment body on IssueCommentEvent Now properly filters by issue body content All tests pass and the implementation builds successfully. The webhook system now properly handles rate limiting, ensures atomic idempotency checks, creates valid owner references for garbage collection, generates valid Kubernetes resource names, and correctly applies all filter conditions. * chore: generate
* Add HTTP timeouts to webhook server to prevent resource exhaustion attacks - ReadTimeout: 30s, WriteTimeout: 30s, ReadHeaderTimeout: 10s, IdleTimeout: 120s - Protects against Slowloris and connection holding attacks * Add ownership verification before deleting stale resources - Prevents webhook reconciliation from deleting unrelated Deployments/CronJobs - Only deletes resources owned by TaskSpawners with proper controller reference * Enable strict template validation with missingkey=error - Missing template variables now surface as errors instead of silent "<no value>" - Prevents malformed prompts/labels/annotations from invalid templates - Applied to both taskbuilder and source template rendering * Update controller tests with proper owner references for stale resource scenarios Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* feat: add Slack type to TaskSpawner API Add Slack as a new source type in the When struct, enabling ChatOps-driven agent execution from Slack messages via Socket Mode. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: remove ResponseChannel from Slack type Bot responses will always be posted as thread replies to the originating message. No override is needed. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: implement SlackSource with Socket Mode listener (#36) * feat: implement SlackSource with Socket Mode listener Add SlackSource that connects to Slack via Socket Mode and accumulates WorkItems from channel messages. Discover() drains the queue each cycle. Event processing logic (shouldProcess, matchesChannel, matchesUser, buildWorkItem) is extracted as pure functions for testability. The Socket Mode goroutine is a thin glue layer. Filtering: ignores threaded replies, bot messages, and self-messages. Optional trigger command prefix support with stripping. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: address review feedback for SlackSource - Replace started bool with sync.Once to fix potential data race - Use context.Background() for Socket Mode goroutine lifetime - Add 5-second timeout for enrichment API calls - Use composite key for slash command work item ID - Filter message_changed, message_deleted, message_replied subtypes - Add test cases for new filtered subtypes Towards AIE-17 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: guard s.cancel with mutex to prevent data race in Stop() s.cancel is written in Start() and read in Stop() — without synchronization this is a race if Stop() is called while Start() is still running inside startOnce.Do. Towards AIE-17 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: wire Slack source into deployment builder and spawner (#37) * feat: wire Slack source into deployment builder and spawner - Add Slack block to buildPodParts() mounting SLACK_BOT_TOKEN and SLACK_APP_TOKEN from SecretRef, passing optional CLI args for trigger command, channels, and allowed users - Add CLI flags and spawnerRuntimeConfig fields for Slack config - Add Slack case to buildSource() and resolvedPollInterval() - Add parseCSV helper for comma-separated CLI flag values - Add TestDeploymentBuilder_Slack and TestDeploymentBuilder_SlackMinimal Towards AIE-17 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * test: add TestBuildSource_Slack for buildSource coverage Verifies the Slack branch in buildSource() correctly reads env vars for bot/app tokens and parses CSV CLI args for trigger command, channels, and allowed users. Towards AIE-17 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: add Slack status feedback as thread replies (#38) * feat: add Slack status feedback as thread replies When a Slack message triggers a Task, the bot now posts thread replies to the originating message as the Task progresses: - "Working on your request..." when accepted - "Done!" when succeeded - "Failed." when failed Implementation: - SlackReporter posts/updates thread replies via slack-go - SlackTaskReporter mirrors GitHubTaskReporter pattern using Slack-specific annotations for channel, thread_ts, reply_ts - sourceAnnotations() stamps Slack metadata on Tasks at creation - reportingEnabled() returns true for Slack sources (always-on) - buildWorkItem() now carries channel ID for annotation stamping Towards AIE-17 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: enrich Slack status messages and address review feedback - Include PR URL in succeeded messages and error details in failed messages - Extract SlackMessenger interface for testability (enables fake injection) - Reuse slack.Client instead of creating one per API call - Validate SLACK_BOT_TOKEN is set before constructing reporter - Add full post+persist and update path tests for SlackTaskReporter Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: skip thread_ts annotation for slash command work items Slash command item IDs are compound strings (e.g. "C123:/cmd:trigger") that are not valid Slack message timestamps. Using them as thread_ts would cause PostThreadReply to fail with invalid_arguments. Only set the annotation when the ID is a valid Slack timestamp (digits.digits). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* docs: move pollInterval to per-source form in README snippets The README YAML snippets used the deprecated top-level spec.pollInterval field. Move pollInterval inside the githubIssues source block in both the self-development section and the TaskSpawner quick-start example, aligning them with the reference docs, examples, and the actual kelos-workers.yaml manifest. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * Update opencode image to 1.3.9 * Update claude-code image to 2.1.88 * Update cursor image to 2026.03.30-a5d3e17 * feat: log cache misses in ghproxy * fix: use fetch-depth 2 for fork e2e merge ref checkout Shallow clone with depth 1 prevents git from reporting parent SHAs of the merge commit, causing the head SHA validation to always fail. * feat: add github webhook events (#34) * feat: add github webhook events * fix: build images * Fixed Issues 1. P1 - MaxConcurrencyError HTTP Response ✅ Fixed ServeHTTP to properly handle MaxConcurrencyError with HTTP 503 and Retry-After header Previously returned generic HTTP 500, now correctly returns 503 for rate limiting 2. P1 - Empty APIVersion/Kind in Owner References ✅ Fixed owner reference creation to use client.Scheme().ObjectKinds() to get proper GVK Added missing Controller and BlockOwnerDeletion fields for proper garbage collection Previously had empty strings, now has correct API version and kind 3. P1 - TOCTOU Race Condition in Idempotency ✅ Replaced separate IsProcessed()/MarkProcessed() with atomic CheckAndMark() Eliminated race window where two requests with same delivery ID could both pass the check Now uses single lock operation for check-and-set 4. P2 - Invalid Kubernetes Names from Truncation ✅ Fixed task name generation to handle nil/empty ID values safely Added strings.TrimRight(taskName[:63], "-.") to ensure names don't end with invalid characters Prevents server-side validation failures 5. P2 - BodyContains Filter Ignored for IssuesEvent ✅ Fixed GitHub filter to check issue body for BodyContains on IssuesEvent Previously only checked comment body on IssueCommentEvent Now properly filters by issue body content All tests pass and the implementation builds successfully. The webhook system now properly handles rate limiting, ensures atomic idempotency checks, creates valid owner references for garbage collection, generates valid Kubernetes resource names, and correctly applies all filter conditions. * chore: generate * fix: normalize the tasknames * feat: scope ghproxy by workspace with own auth Replace the centralized static ghproxy with per-workspace instances managed by a new WorkspaceReconciler. Each workspace ghproxy owns its GitHub auth directly via token-refresher sidecar, fixing the ETag-per-token cache invalidation cycle caused by multiple spawner pods using different tokens through a shared proxy. Also logs cache misses with the cache key for observability. * fix: populate SpawnerRef fields for Task owner references PR kelos-dev#851 introduced taskbuilder.BuildTask with SpawnerRef but the spawner call site only set Name, leaving APIVersion, Kind, and UID empty. Kubernetes rejects owner references with empty required fields, causing Task creation to fail. Also remove BlockOwnerDeletion since neither the spawner nor webhook RBAC role has the permissions Kubernetes requires for that field. * Merge upstream prod into fork: add GitHub webhook improvements, Linear webhooks, Slack support Merges upstream GitHub webhook events and Linear webhook support into the fork which already has Slack integration. All three webhook sources (GitHub, Linear, Slack) are now supported alongside cron and poll-based task spawners. Resolves all merge conflicts across 21 files. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Gunju Kim <gjkim042@gmail.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* Fix SlackSource being rebuilt every poll cycle * Add test * Use CRD source type instead of env vars * Add retry with backoff for TaskSpawner CRD fetch at startup * Fix test
Add when.linearWebhook source type that triggers task creation from Linear webhook events. Includes HMAC-SHA256 signature validation, filtering by resource type, action, workflow state, and labels, and Helm chart support for the Linear webhook server alongside GitHub.
* feat: handle slack responses properly * fix: review comments
* docs: add contributing guide and recommended steps * Update ANOMALO-CONTRIBUTING-GUIDE.md Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> --------- Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
* feat: Add support for fileChanged Patterns * Update internal/webhook/handler.go Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * fix: review comments --------- Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
* fix: improve Slack thread interactivity for gravity agent Thread replies were silently dropped by shouldProcess(), preventing follow-up conversations. Now thread replies create new tasks with the full thread history as context via GetConversationRepliesContext. Status messages were edited in-place, causing confusing "(edited)" tags. Now each phase change posts a new thread reply instead of updating the existing one. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: only process thread replies where bot participated Address greptile review feedback: - Add botParticipated() check so thread replies in unrelated conversations don't spawn tasks - Use a separate 10s timeout for GetConversationRepliesContext instead of sharing the 5s enrichCtx across all API calls Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat: use Slack Block Kit for rich message formatting Replace plain text Slack replies with Block Kit blocks for a more human-friendly experience. Status messages now use emoji headers, structured sections for agent responses and PR links, and a muted context block for the task name. Fallback text is preserved for notifications and accessibility. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * Address Greptile comments * Fix codestyle --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Linear Comment webhooks include data.issue.id/title/url but not
data.issue.description. This causes Go template rendering to crash
with "map has no entry for key description" when spawner prompt
templates reference {{.Payload.data.issue.description}}.
Extend the existing label enrichment GraphQL query to also fetch the
issue description, and make enrichment unconditional for all Comment
events (not gated on spawner filter needs). This prevents a class of
template errors cheaply with a single API call that fetches both
labels and description.
Co-authored-by: Hans Knecht <hans@anomalo.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Docs: add commentPolicy, reporting, and excludeAuthors to reference
…56) #### What type of PR is this? /kind bug #### What this PR does / why we need it: Slack's `mrkdwn` text format does not support markdown tables, lists, or headers. When an agent response contains these, they render as garbled plain text in Slack messages. This PR adds a markdown-to-Block Kit parser (`responseToBlocks`) that splits the decoded agent response into segments and converts each to the appropriate Slack block type: - Markdown tables (`| col | col |`) become `TableBlock` with `RichTextBlock` cells - Unordered lists (`- item`, `* item`) and ordered lists (`1. item`) become `RichTextBlock` with `RichTextList` - Headers (`#`, `##`, `###`) become `HeaderBlock` - Plain text remains a `SectionBlock` with `mrkdwn` (unchanged behavior) `FormatSlackSucceeded` and `FormatSlackFailed` now use `responseToBlocks()` instead of placing the entire response into a single `SectionBlock`. #### Which issue(s) this PR is related to: N/A #### Special notes for your reviewer: - All block types use the existing `slack-go/slack` v0.20.0 library (`TableBlock`, `RichTextBlock`, `HeaderBlock`) -- no custom structs needed. - The parser handles edge cases: `*bold*` is not falsely matched as a list item, `3.14` is not matched as an ordered list, and pipe characters in plain text don't become tables unless followed by a separator row. - Slack limits tables to 100 rows and 1 table per message. We don't enforce this client-side -- invalid input will surface as a Slack API error. - Nested/indented lists are not supported; all list items must be at the same level. #### Does this PR introduce a user-facing change? ```release-note Slack messages now render markdown tables, lists, and headers using native Block Kit blocks instead of raw text. ```
#### What type of PR is this?
/kind bug
#### What this PR does / why we need it:
Fixes `invalid_blocks` error from the Slack API that prevented all Slack messages from being posted.
The `tableBlock()` function in `slack_blocks.go` was setting `column_settings` with an empty `align` field (`"align": ""`). Slack only accepts `"left"`, `"center"`, or `"right"` — an empty string causes the entire message to be rejected with `invalid_blocks`. Since `column_settings` is optional, the fix removes it entirely.
#### Which issue(s) this PR is related to:
N/A
#### Special notes for your reviewer:
- Root cause: `slack.ColumnSetting{IsWrapped: true}` leaves `Align` as its zero value (`""`), which `encoding/json` serializes as `"align": ""`. The Slack API rejects this.
- The `slack-go` library's own `TableBlock` test also uses no `column_settings`, confirming they're optional.
- The error was retried every ~10 seconds per reconcile loop, visible in spawner logs as `posting Slack thread reply: invalid_blocks`.
#### Does this PR introduce a user-facing change?
```release-note
Fix Slack message posting failure caused by invalid table block column settings.
```
…Spawner` deployments (#51) /kind feature Introduces a centralized `kelos-slack-server` binary that replaces the per-`TaskSpawner` Slack deployment model. Instead of each `TaskSpawner` with a `spec.when.slack` source running its own spawner pod with a dedicated Socket Mode WebSocket connection, a single server now handles all Slack events and routes them to matching `TaskSpawners` via the Kubernetes API — following the same pattern as `kelos-webhook-github` and `kelos-webhook-linear`. **Why this matters:** - Slack App Tokens support only one concurrent Socket Mode connection. The per-`TaskSpawner` model breaks when multiple `TaskSpawners` share the same Slack app. - Reduces resource overhead — one pod instead of N pods for N Slack-enabled `TaskSpawners`. - Aligns Slack with the existing webhook architecture, making the codebase more consistent. **What's included:** | Path | Description | |------|-------------| | `cmd/kelos-slack-server/main.go` | Server binary using `controller-runtime` for health probes, metrics, leader election, and a reporting loop | | `internal/slack/handler.go` | Central `SlackHandler` — connects via Socket Mode, dispatches `EventsAPI` and `SlashCommand` events, routes messages to matching `TaskSpawners` | | `internal/slack/filter.go` | `SlackMessageData` struct, channel/user matching (`MatchesSpawner`), trigger command processing (`ProcessTriggerCommand`), and template variable extraction (`ExtractSlackWorkItem`) | | `internal/slack/thread.go` | Thread context fetching (`FetchThreadContext`), bot participation check (`BotParticipated`), and conversation formatting (`FormatThreadContext`) | | `internal/slack/filter_test.go` | Tests for matching, trigger processing, work item extraction, and `shouldProcess` | | `internal/slack/thread_test.go` | Tests for `BotParticipated` and `FormatThreadContext` | | `internal/controller/taskspawner_controller.go` | Added `Slack` to `isWebhookBased` so the controller skips spawner-pod creation for Slack `TaskSpawners` | | `api/v1alpha1/taskspawner_types.go` | Removed `SecretRef` and `PollInterval` from `Slack` struct — these are now server-side concerns | | `internal/controller/taskspawner_deployment_builder.go` | Removed Slack args/env injection from `buildPodParts` | | `cmd/kelos-spawner/main.go` | Removed all Slack flags, persistent source, reporting, and `buildSourceWithProxy` Slack branch | | `cmd/kelos-spawner/reconciler.go` | Removed Slack config fields, reporting branch, and poll interval case | | `internal/source/slack.go` | Deleted — per-`TaskSpawner` Slack source is no longer needed | | `internal/source/slack_test.go` | Deleted | | `internal/manifests/charts/kelos/templates/slack-server.yaml` | Helm template for the centralized `kelos-slack-server` `Deployment` | | `internal/manifests/charts/kelos/templates/rbac.yaml` | Added `kelos-slack-server-role` `ClusterRole` and binding | | `internal/manifests/charts/kelos/templates/serviceaccount.yaml` | Added `kelos-slack-server` `ServiceAccount` | | `internal/manifests/charts/kelos/values.yaml` | Added `slackServer` values (`enabled`, `replicas`, `secretName`, `image`, `resources`) | | `internal/manifests/charts/kelos/templates/crds/taskspawner-crd.yaml` | Removed `secretRef` and `pollInterval` from Slack schema, updated description | | `internal/manifests/install-crd.yaml` | Same CRD changes as above (standalone install manifest) | **Message routing flow:** 1. Socket Mode receives a Slack event 2. `handleEventsAPI` / `handleSlashCommand` parse and enrich the message 3. `routeMessage` lists all `TaskSpawners` with `spec.when.slack` set 4. For each spawner: checks suspended state, `maxConcurrency`, channel/user filters, and trigger command 5. Creates a `Task` with Slack reporting annotations (`kelos.dev/slack-reporting`, `kelos.dev/slack-channel`, `kelos.dev/slack-thread-ts`) N/A - The reporting loop in `main.go` lists **all** `Tasks` cluster-wide (not scoped to a single `TaskSpawner`), matching the centralized model. Each `Task`'s Slack annotations determine whether reporting applies. - Thread interactivity: thread replies only trigger tasks if the bot has previously participated in the thread (`BotParticipated` check), preventing noise from unrelated conversations. - The Helm chart follows the same pattern as `webhook-server.yaml` — `Deployment` with `ServiceAccount`, `ClusterRole`, and `ClusterRoleBinding`. The Slack server additionally needs `update` on `tasks` (for reporting annotations) and `leases`/`events` (for leader election). ```release-note Add centralized `kelos-slack-server` for routing Slack messages to matching `TaskSpawners`, replacing the per-`TaskSpawner` Socket Mode deployment model. Remove `secretRef` and `pollInterval` from the `Slack` CRD struct. Add Helm chart support via `slackServer.enabled`. ```
…erver` image (#60) /kind bug The `kelos-slack-server` binary was added in #59 but was missing a `Dockerfile` and wasn't included in the `IMAGE_DIRS` list in the `Makefile`. CI wouldn't build or push the `kelos-slack-server` Docker image, so it can't be deployed. This adds: - `cmd/kelos-slack-server/Dockerfile` (same `distroless/static:nonroot` pattern as the other server images) - `cmd/kelos-slack-server` to `IMAGE_DIRS` in the `Makefile` N/A The `Dockerfile` is identical in structure to `cmd/kelos-webhook-server/Dockerfile`. ```release-note NONE ```
* feat: edit Slack progress messages in-place instead of posting new ones Progress updates from agent pod logs now edit a single thread reply rather than posting a new message each time the text changes. Status transition messages (accepted, succeeded, failed) continue to post as new thread replies. Changes: - Add UpdateThreadReply to SlackMessenger interface and SlackReporter - Track per-task progress message timestamp in progressTS map - First progress update posts a new reply; subsequent updates edit it - Clean up progressTS alongside lastProgress on terminal phase and sweep Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: clear stale progressTS on update failure When a Slack progress message is deleted externally, UpdateMessage fails on every tick but progressTS was never cleared, preventing fallback to posting a fresh reply. Clear the stale TS on error so the next tick recovers by posting a new message. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Gravity Bot <gravity@anomalo.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> Co-authored-by: Jasmine Ahuja <jasmine@anomalo.com>
…#96) * feat(webhook): add extraEnv to generic webhook for per-source secrets Allow mounting individual Kubernetes secrets as env vars in the generic webhook server pod. This supports the <SOURCE>_WEBHOOK_SECRET convention when each source has its own secret (e.g., linear-comment-webhook-secret). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: uncomment generic webhook example in helm-values-webhook.yaml Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: address review feedback on generic webhook extraEnv Remove redundant outer `if or` guard in the Helm template — each inner block already gates itself. Assert `key: WEBHOOK_SECRET` in the test to catch template regressions on the secretKeyRef key field. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
#97) * feat(deploy): enable generic webhook with Linear comment secret in dev Wire up the generic webhook server in the dev deployment values, injecting LINEAR_COMMENT_WEBHOOK_SECRET from the linear-comment-webhook-secret Kubernetes Secret via extraEnv. Add rollout restart/status for the new kelos-webhook-generic deployment. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat(deploy): add PodMonitoring for generic webhook server Add a PodMonitoring resource for kelos-webhook-generic so its metrics are scraped, consistent with the existing kelos-webhook-github monitor. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Add a new ChannelTypes field to the Slack spawner config that filters by Slack channel type (channel, group, im, mim). This enables spawners that only fire on DMs, so users can message the bot without @-mentioning it. For slash commands, "directmessage" is mapped to "im" for consistency. Empty list means match all (backward compatible). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
github-actions
Bot
added
needs-triage
needs-kind
needs-priority
needs-actor
needs-release-note
labels
knechtionscoding
force-pushed
the
prod
branch
4 times, most recently
from
May 1, 2026 17:47
b9dd057 to
d9ddb7c
Compare
knechtionscoding
force-pushed
the
prod
branch
2 times, most recently
from
June 11, 2026 11:13
0bb9f5d to
7dc52bc
Compare
knechtionscoding
force-pushed
the
prod
branch
4 times, most recently
from
July 10, 2026 13:46
f27e003 to
45a666a
Compare
knechtionscoding
force-pushed
the
prod
branch
2 times, most recently
from
July 23, 2026 14:25
2b3ee43 to
e5a56d9
Compare
knechtionscoding
force-pushed
the
prod
branch
2 times, most recently
from
July 28, 2026 14:13
4797eb0 to
1bccf1a
Compare
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
6 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.
Summary
channelTypesfield to theSlackTaskSpawner config that filters by Slack channel type (channel,group,im,mim)@-mentions — users can just type in a DMchannelName: "directmessage"to"im"for consistent filteringchannelTypeslist matches all channel types (fully backward compatible)Context
Companion to https://github.com/datagravity-ai/dquality/pull/27735, which adds the DM spawner YAML configs referencing this field. The config PR is inert until this Kelos change lands.
Changes
api/v1alpha1/taskspawner_types.goChannelTypes []stringfield toSlackstructapi/v1alpha1/zz_generated.deepcopy.gointernal/slack/filter.goChannelTypetoSlackMessageData, addmatchesChannelTypehelper, integrate intoMatchesSpawnerinternal/slack/handler.goChannelTypefrom message events and slash commandsinternal/slack/filter_test.gomatchesChannelType,MatchesSpawnerwith channelTypes, and full DM routing scenarioTest plan
TestMatchesChannelTypeunit tests cover match/reject/empty-list casesTestDMRoutingintegration test validates DM vs channel spawner routinggo build ./...succeeds@-mention and don't hit DM spawners🤖 Generated with Claude Code