fix: AI channel blocklist - rebase on main by BillChirico · Pull Request #192 · VolvoxLLC/volvox-bot

BillChirico · 2026-03-02T01:51:31Z

Rebased #187 on latest main to resolve BEHIND status.

This is a simple rebase fix - the original PR adds an config option to control which channels the AI assistant is allowed to respond in.

Changes:

Added array to config section
Updated config validation schema
Updated dashboard config editor with channel picker
Updated TypeScript types

…itor

coderabbitai · 2026-03-02T01:51:40Z

Warning

Rate limit exceeded

@BillChirico has exceeded the limit for the number of commits that can be reviewed per hour. Please wait 11 minutes and 29 seconds before requesting another review.

⌛ How to resolve this issue?

After the wait time has elapsed, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout.

Please see our FAQ for further information.

📥 Commits

Reviewing files that changed from the base of the PR and between 11a9bb5 and 648caff.

📒 Files selected for processing (2)

config.json
src/modules/events.js

📝 Walkthrough

Walkthrough

This PR introduces an AI feedback tracking system with channel blocklist enforcement. It adds database infrastructure for recording user reactions to AI messages, API routes for retrieving feedback statistics, event handlers to capture feedback via reactions, and UI components for visualization and configuration management.

Changes

Cohort / File(s)	Summary
Configuration Schema `config.json`, `src/api/utils/configValidation.js`, `web/src/types/config.ts`	Extended AI configuration with feedback tracking fields (channels, historyLength, historyTTLDays, blockedChannelIds, feedback.enabled); normalized emoji badge labels to Unicode escape sequences.
Database Migration `migrations/003_ai_feedback.cjs`	New migration creating ai_feedback table with serial ID, message/channel/guild/user IDs, feedback type (positive/negative), timestamps, unique constraint per message-user pair, and indexes on guild_id and message_id.
AI Feedback Module `src/modules/aiFeedback.js`	New module providing in-memory AI message tracking with LRU eviction (2000 capacity), database operations for recording/deleting feedback, and aggregated statistics/trend queries with optional test pool injection.
AI Channel Blocklist `src/modules/ai.js`	Added isChannelBlocked function to determine if a channel or its parent thread is in the guild's blocklist, with fail-open error handling.
Event Integration `src/modules/events.js`, `src/modules/triage-respond.js`	Integrated channel blocklist checks to prevent AI responses in blocked channels; added reaction handlers to record/delete feedback on AI message reactions; added automatic feedback emoji reactions to AI responses when enabled.
API Endpoints `src/api/index.js`, `src/api/routes/ai-feedback.js`	New protected API endpoint mounted at /guilds/:id/ai-feedback with GET /stats (aggregated feedback and daily trends) and GET /recent (recent feedback entries) routes, both with query parameter validation and rate limiting.
Web UI Components `web/src/components/dashboard/ai-feedback-stats.tsx`, `web/src/components/dashboard/config-editor.tsx`	New feedback stats dashboard displaying positive/negative counts, approval ratio, and charts; extended config editor with ChannelSelector for managing blocked channels.
Test Suite `tests/api/routes/ai-feedback.test.js`, `tests/modules/ai.test.js`, `tests/modules/aiFeedback.test.js`, `tests/modules/events.test.js`	Comprehensive unit and integration tests covering API routes, channel blocklist logic, feedback module operations, and event handler integration with error cases and edge conditions.

Possibly Related PRs

PR #190: Directly implements the same AI feedback tracking feature across modules, API routes, and test coverage
PR #175: Adds ChannelSelector UI components that are reused for the blocked channels configuration interface
PR #152: Modifies config.json AI settings and normalizes activity badge emoji values similarly

Suggested Reviewers

Copilot

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'fix: AI channel blocklist - rebase on main' clearly summarizes the main change: rebasing an AI channel blocklist feature onto main to resolve being behind.
Description check	✅ Passed	The description explains the rebase context (PR `#187`), mentions the AI channel blocklist feature, and lists concrete changes related to config, validation, dashboard, and TypeScript types, all of which are present in the changeset.
Docstring Coverage	✅ Passed	Docstring coverage is 90.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch fix/pr-187-resolved

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Copilot

Pull request overview

Rebases and carries forward the original feature to add a per-guild AI channel blocklist, preventing the AI assistant from responding (and accumulating triage context) in configured channels and their threads.

Changes:

Adds ai.blockedChannelIds to config types, validation schema, and default config.json.
Introduces isChannelBlocked() and enforces the blocklist in the message-create event handler (including thread-parent blocking).
Extends the dashboard config editor with a channel picker for blocked channels and adds/updates unit tests.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
`src/modules/ai.js`	Adds `isChannelBlocked()` helper to evaluate blocklist membership (including thread parent checks).
`src/modules/events.js`	Applies blocklist gating in the AI message handling path.
`src/api/utils/configValidation.js`	Extends config schema to allow `ai.blockedChannelIds`.
`web/src/types/config.ts`	Updates TypeScript config types to include `blockedChannelIds`.
`web/src/components/dashboard/config-editor.tsx`	Adds a “Blocked Channels” UI section using `ChannelSelector`.
`tests/modules/ai.test.js`	Adds unit tests for `isChannelBlocked()`.
`tests/modules/events.test.js`	Adds event-handler tests to ensure blocked channels/threads suppress AI behavior.
`config.json`	Adds default `ai.blockedChannelIds` and updates seed config content formatting/encoding.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

src/modules/events.js

src/modules/ai.js

tests/modules/ai.test.js

config.json

greptile-apps · 2026-03-02T02:01:41Z

Greptile Summary

This PR rebases the AI channel blocklist feature from #187 onto main, adding a blockedChannelIds configuration option that prevents the AI assistant from responding in specified channels.

Added blockedChannelIds: string[] to AI config with schema validation and TypeScript types
Implemented isChannelBlocked() function with proper error handling (fails open when config unavailable)
Integrated blocklist check in message handler; blocklist takes precedence over allowlist
For threads, blocking a parent channel automatically blocks all child threads
Added comprehensive test coverage for both the blocking function and event handler integration
Dashboard UI includes new channel selector for configuring blocked channels

The implementation follows existing code patterns, handles edge cases appropriately, and includes thorough test coverage.

Confidence Score: 5/5

This PR is safe to merge with minimal risk
Clean rebase of well-tested feature addition. Implementation includes proper error handling, comprehensive test coverage, and follows existing patterns. No bugs or security issues identified. All validation, TypeScript types, and UI components are correctly implemented.
No files require special attention

Important Files Changed

Filename	Overview
src/api/utils/configValidation.js	Added schema validation for `blockedChannelIds` array
src/modules/ai.js	Implemented `isChannelBlocked` with proper error handling and fail-open behavior
src/modules/events.js	Integrated blocklist check in message handler, blocks AI responses in configured channels
web/src/components/dashboard/config-editor.tsx	Added UI card with `ChannelSelector` for configuring blocked channels

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[Message Received] --> B{AI Enabled?}
    B -->|No| Z[Ignore]
    B -->|Yes| C{Mentioned or Reply?}
    C -->|No| D[Accumulate for Context]
    C -->|Yes| E{In Allowed Channels?}
    E -->|No| Z
    E -->|Yes| F{Channel Blocked?}
    F -->|Yes| Z
    F -->|No| G[Accumulate Message]
    G --> H[Show Typing Indicator]
    H --> I[Evaluate & Respond]
    
    style F fill:#f96,stroke:#333,stroke-width:2px
    style Z fill:#ddd

_{Last reviewed commit: 648caff}

src/api/index.js

src/api/routes/ai-feedback.js

+router.get(
+  '/stats',
+  feedbackRateLimit,
+  requireGuildAdmin,


src/api/routes/ai-feedback.js

+router.get(
+  '/recent',
+  feedbackRateLimit,
+  requireGuildAdmin,


coderabbitai

Actionable comments posted: 13

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@config.json`:
- Around line 13-16: Add documentation for the new AI config keys introduced in
config.json: document ai.blockedChannelIds (array of channel IDs to ignore) and
ai.feedback.enabled (boolean to toggle feedback) in the README.md config
reference section; update the config reference table/section to include key
names, types, default values, and a short description for ai.blockedChannelIds
and ai.feedback.enabled so readers can discover and understand these new
settings.

In `@migrations/003_ai_feedback.cjs`:
- Line 19: The migration defines the feedback timestamp column as created_at
TIMESTAMP NOT NULL DEFAULT NOW(); update the column to be timezone-aware by
changing its type to TIMESTAMPTZ (i.e., created_at TIMESTAMPTZ NOT NULL DEFAULT
NOW()) so feedback event times are stored with timezone information; locate the
created_at definition in migrations/003_ai_feedback.cjs and replace TIMESTAMP
with TIMESTAMPTZ (ensure any downstream code that assumes TIMESTAMP handles
timestamptz values).
- Around line 24-39: Add a composite index on (guild_id, created_at) to speed
guild-scoped time queries: in the migration's exports.up where pgm.sql creates
the two single-column indexes for ai_feedback, add a pgm.sql call to CREATE
INDEX IF NOT EXISTS idx_ai_feedback_guild_id_created_at ON ai_feedback(guild_id,
created_at); and in exports.down add a corresponding DROP INDEX IF EXISTS
idx_ai_feedback_guild_id_created_at; so the migration creates and removes the
composite index alongside the existing idx_ai_feedback_guild_id and
idx_ai_feedback_message_id.

In `@src/api/routes/ai-feedback.js`:
- Around line 103-135: The route handler currently sends responses directly;
instead import and throw the app's custom errors (e.g., ServiceUnavailableError
and InternalServerError from src/utils/errors.js) so the centralized error
middleware enforces the API contract. Replace the manual return
res.status(503)... when getDbPool(req) is falsy with: throw new
ServiceUnavailableError('Database not available') and in the catch block call
logError('Failed to fetch AI feedback stats', { guildId: req.params.id, error:
err.message, err }) then re-throw or throw new InternalServerError('Failed to
fetch AI feedback stats', { cause: err }) so the global error handler formats
the response; apply the same pattern to the other handler referenced (lines
208-233) and remove direct res.status/res.json usage for error cases.

In `@src/api/utils/configValidation.js`:
- Line 20: The schema entry for blockedChannelIds is too permissive and should
only accept an array of channel ID strings; update the config validation schema
(the blockedChannelIds property in src/api/utils/configValidation.js) to enforce
items: { type: 'string' } and add uniqueItems: true (and optionally a pattern or
minLength if you need numeric/string format constraints) so invalid element
types no longer persist and block matching remains reliable.

In `@src/modules/aiFeedback.js`:
- Around line 177-181: The code sets negative = row?.negative || 1 which
incorrectly treats a legitimate zero as one; change the fallback to preserve
zero by using a nullish/coalescing check (e.g., with ??) or explicit undefined
check so negative falls back to 0 (or the intended default) when row?.negative
is undefined, updating the assignment for negative in the block that reads
result.rows[0] (referencing row and negative in src/modules/aiFeedback.js).

In `@src/modules/events.js`:
- Around line 224-228: Move the channel blocklist guard earlier inside the AI
handling branch so it runs before any reply-reference or message-fetch logic;
specifically, compute parentId (using message.channel.isThread?.() and
message.channel.parentId) and call isChannelBlocked(message.channel.id,
parentId, message.guild.id) as the first early-return check in the AI path to
avoid performing any Discord fetches when the channel is blocked.
- Around line 355-366: The current handler calls deleteFeedback unconditionally
when a feedback reaction is removed; change it to only delete when the user has
no remaining feedback reactions on that message. Specifically, in the
reaction-remove branch (where guildConfig.ai?.feedback?.enabled,
isAiMessage(reaction.message.id), FEEDBACK_EMOJI and deleteFeedback are used),
determine the opposite emoji (FEEDBACK_EMOJI.positive vs .negative) and check
the message's reactions/users to see if the user still has the opposite reaction
(use reaction.message.reactions.cache or fetch the reaction users and test for
user.id); only call deleteFeedback({ messageId: reaction.message.id, userId:
user.id }) when the user does not have either feedback reaction remaining.

In `@tests/api/routes/ai-feedback.test.js`:
- Around line 176-194: The mocked DB rows in the test for the
/api/v1/guilds/:id/ai-feedback/recent route are using snake_case keys, but
getRecentFeedback (and the route handler) returns camelCase fields; update the
fakeRows fixture in tests/api/routes/ai-feedback.test.js to use camelCase keys
(e.g., messageId, channelId, userId, feedbackType, createdAt) so the mocked
query shape matches the module's output and the existing assertions (messageId,
feedbackType) remain valid.

In `@tests/modules/ai.test.js`:
- Around line 210-257: Add a test that exercises isChannelBlocked with a guildId
to verify guild-scoped configs: mock getConfig to return { ai: {
blockedChannelIds: [...] } } when called with that guildId, call
isChannelBlocked(channelId, parentId, 'guild-123'), assert the returned boolean
matches expectation (true/false based on blockedChannelIds), and assert
getConfig was called with 'guild-123' (e.g.,
expect(getConfig).toHaveBeenCalledWith('guild-123')). Ensure the test references
isChannelBlocked, getConfig, and blockedChannelIds so guild-specific fallback
regressions are detected.

In `@tests/modules/aiFeedback.test.js`:
- Around line 59-178: Add unit tests for deleteFeedback mirroring other suites:
(1) "does nothing when no pool is configured" — call deleteFeedback({ messageId,
channelId, guildId, userId }) without setting a pool and assert mockPool.query
was not called; (2) "deletes feedback via pool query" — setPool(mockPool) with
mockPool.query.mockResolvedValue({ rows: [] }), call deleteFeedback(...) and
assert mockPool.query was called with a SQL string containing 'DELETE FROM
ai_feedback' and the correct params [messageId, channelId, guildId, userId]; (3)
"handles DB errors gracefully without throwing" — setPool(mockPool) with
mockPool.query.mockRejectedValue(new Error('DB error')), call
deleteFeedback(...) and assert the promise resolves to undefined (does not
throw). Optionally add a test using _setPoolGetter(() => altPool) to validate DI
like other tests.

In `@web/src/components/dashboard/ai-feedback-stats.tsx`:
- Around line 45-71: fetchStats can suffer from stale responses overwriting
state when selectedGuild changes; modify fetchStats/useEffect to cancel or
ignore out-of-date requests by using an AbortController or a per-request id.
Specifically, in the effect that calls fetchStats, create an AbortController (or
increment a local requestId counter) and pass its signal (or id) into
fetchStats; inside fetchStats check the signal.aborted (or compare requestId
before calling setStats/setLoading/setError) and skip state updates if
cancelled/mismatched; also ensure fetch is called with the controller.signal and
that you call controller.abort() (or discard the id) in the effect cleanup so
only the latest guild’s response can update stats.

In `@web/src/components/dashboard/config-editor.tsx`:
- Around line 771-779: Replace the direct-fetch ChannelSelector usage with the
dashboard's Zustand-backed channel source: read the cached channels for guildId
from the discord-entities store (e.g., useDiscordEntitiesStore or
discordEntitiesStore selector that returns channels for a guild) and pass that
channel list into ChannelSelector instead of allowing it to perform its own
fetch; keep the same props (id "ai-blocked-channels",
selected={(draftConfig.ai?.blockedChannelIds ?? []) as string[]},
onChange={updateAiBlockedChannels}, placeholder, disabled={saving},
filter="text") but supply the store-backed channels via the component’s
channels/options prop (or a source prop used by our other selectors) so
ChannelSelector uses the cached Zustand data for guildId.

ℹ️ Review info

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between f53e370 and 11a9bb5.

📒 Files selected for processing (16)

config.json
migrations/003_ai_feedback.cjs
src/api/index.js
src/api/routes/ai-feedback.js
src/api/utils/configValidation.js
src/modules/ai.js
src/modules/aiFeedback.js
src/modules/events.js
src/modules/triage-respond.js
tests/api/routes/ai-feedback.test.js
tests/modules/ai.test.js
tests/modules/aiFeedback.test.js
tests/modules/events.test.js
web/src/components/dashboard/ai-feedback-stats.tsx
web/src/components/dashboard/config-editor.tsx
web/src/types/config.ts

📜 Review details

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

GitHub Check: Greptile Review

🧰 Additional context used

📓 Path-based instructions (9)

**/*.js