feat(frontend): voice input as an experimental per-user settings flag

[mmabrouk]

Voice input (dictation in the agent chat composer) becomes a per-user experimental toggle in Settings → Feature flags, default off, persisted per user like the neighboring flags. The env variable stays as a dev-stack override (either switch turns the feature on). Nothing else about the recorder or its interplay with attachments changes.

Before: voice input is invisible unless the deployment bakes NEXT_PUBLIC_AGENT_VOICE_INPUT=true into the frontend build; users have no way to try it.
After: any user can turn it on for themselves under Settings → Feature flags → "Voice input"; it stays off for everyone else.

New user-visible strings, complete inventory: the flag label "Voice input" and its description "Dictate messages in the agent chat."

Covered by unit tests for the availability combinator (setting off + env off, setting on, env on). Verified with tsgo and the web unit suite (23 files, 145 tests green).

Intended for the next release. Trial plan: deploy this branch to staging, turn the flag on for a user, run the dictation QA pass.

https://claude.ai/code/session_01McMogkcDRV7UpSAjfd8VKG

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
agenta-documentation	Ready	Preview	Aug 2, 2026 6:19pm

[coderabbitai]

📝 Walkthrough

Summary by CodeRabbit

• New Features

  • Added a Voice input toggle in Feature Flags settings.
  • Enabled dictation in agent chat when voice input is turned on.
  • Voice input can also be enabled through an environment configuration flag.

• Bug Fixes

  • Voice input is now disabled by default unless explicitly enabled.

Walkthrough

The change adds a persisted user setting for agent voice input, exposes a settings toggle, and combines that setting with the NEXT_PUBLIC_AGENT_VOICE_INPUT environment flag when the voice composer checks availability.

Changes

Agent voice input

Layer / File(s)	Summary
Persisted voice input setting web/oss/src/state/settings/featureFlags.ts	Adds agentVoiceInputEnabledAtom with per-user persistence and disabled behavior when no user is available.
Availability helper and tests web/oss/src/components/AgentChatSlice/assets/constants.ts, web/oss/src/components/AgentChatSlice/assets/constants.test.ts	Adds isAgentVoiceInputAvailable and tests availability through the user setting and environment flag.
Settings and composer integration web/oss/src/components/pages/settings/FeatureFlags/FeatureFlags.tsx, web/oss/src/components/AgentChatSlice/hooks/useVoiceComposer.ts	Adds the “Voice input” toggle and uses the atom value when the composer checks availability.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Sequence Diagram(s)

sequenceDiagram
  participant User
  participant FeatureFlags
  participant agentVoiceInputEnabledAtom
  participant useVoiceComposer
  participant isAgentVoiceInputAvailable

  User->>FeatureFlags: Enable Voice input
  FeatureFlags->>agentVoiceInputEnabledAtom: Store setting
  useVoiceComposer->>agentVoiceInputEnabledAtom: Read setting
  useVoiceComposer->>isAgentVoiceInputAvailable: Check setting and environment flag
  isAgentVoiceInputAvailable-->>useVoiceComposer: Return availability

Loading

Possibly related PRs

• Agenta-AI/agenta#5458: Adds the voice-input functionality extended by this setting and availability gating.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title clearly summarizes the main change: adding voice input as an experimental per-user settings flag.
Description check	✅ Passed	The description accurately explains the per-user toggle, default state, environment override, tests, and validation for the changeset.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/voice-input-experimental-flag

---

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

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 1

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: f57c8664-18d0-49fc-895d-e7f3723fc4c7

📥 Commits

Reviewing files that changed from the base of the PR and between ac6b154 and 52ef088.

📒 Files selected for processing (5)

• web/oss/src/components/AgentChatSlice/assets/constants.test.ts
• web/oss/src/components/AgentChatSlice/assets/constants.ts
• web/oss/src/components/AgentChatSlice/hooks/useVoiceComposer.ts
• web/oss/src/components/pages/settings/FeatureFlags/FeatureFlags.tsx
• web/oss/src/state/settings/featureFlags.ts

[github-actions]

Railway Preview Environment

Status	Destroyed (PR closed)

Updated at 2026-08-03T16:42:39.122Z