Add Telegram progress preview flows

[Patrick-Erichsen]

Summary

• add a generic dev channel-message flow runner with Telegram support and thinking-final / working-final preview flows
• make user-facing reasoning/progress labels render as Thinking and Working, with blank-line-separated progress details and no label animation
• route the Telegram working-final flow through native sendMessageDraft while keeping the preview fixture aligned with production static-label behavior
• tighten formatted reasoning detection so visible Thinking... replies are not swallowed, with Discord/Feishu/Telegram parser follow-through

Verification

• git diff --check
• node scripts/run-vitest.mjs test/scripts/channel-message-flows.test.ts
• node scripts/run-vitest.mjs extensions/telegram/src/bot-message-dispatch.test.ts test/scripts/channel-message-flows.test.ts
• node scripts/run-vitest.mjs src/plugin-sdk/reply-payload.test.ts src/agents/tools/message-tool.test.ts extensions/telegram/src/reasoning-lane-coordinator.test.ts test/scripts/channel-message-flows.test.ts --run
• node scripts/run-vitest.mjs extensions/discord/src/monitor/message-handler.process.test.ts extensions/discord/src/chunk.test.ts src/plugin-sdk/channel-streaming.test.ts src/infra/heartbeat-runner.returns-default-unset.test.ts extensions/feishu/src/reply-dispatcher.test.ts extensions/mattermost/src/mattermost/reply-delivery.test.ts extensions/whatsapp/src/auto-reply/deliver-reply.test.ts --run
• AUTOREVIEW_AUTO_TESTS=0 .agents/skills/autoreview/scripts/autoreview --mode branch --base origin/main --reviewer codex --fallback-reviewer none

Real behavior proof

Behavior addressed: Telegram progress previews can be locally exercised with flow fixtures, native Telegram tool-progress previews use sendMessageDraft, and progress/reasoning labels now render as static Working / Thinking labels.
Real environment tested: Local source checkout using the repo Vitest wrapper and autoreview.
Exact steps or command run after this patch: git diff --check; node scripts/run-vitest.mjs test/scripts/channel-message-flows.test.ts; node scripts/run-vitest.mjs extensions/telegram/src/bot-message-dispatch.test.ts test/scripts/channel-message-flows.test.ts; node scripts/run-vitest.mjs src/plugin-sdk/reply-payload.test.ts src/agents/tools/message-tool.test.ts extensions/telegram/src/reasoning-lane-coordinator.test.ts test/scripts/channel-message-flows.test.ts --run; node scripts/run-vitest.mjs extensions/discord/src/monitor/message-handler.process.test.ts extensions/discord/src/chunk.test.ts src/plugin-sdk/channel-streaming.test.ts src/infra/heartbeat-runner.returns-default-unset.test.ts extensions/feishu/src/reply-dispatcher.test.ts extensions/mattermost/src/mattermost/reply-delivery.test.ts extensions/whatsapp/src/auto-reply/deliver-reply.test.ts --run; AUTOREVIEW_AUTO_TESTS=0 .agents/skills/autoreview/scripts/autoreview --mode branch --base origin/main --reviewer codex --fallback-reviewer none.
Evidence after fix: Focused touched-surface runs passed, including 86 Telegram flow/dispatch tests, 180 reasoning/flow tests, and 232 cross-channel reasoning/progress tests; final autoreview reported autoreview clean: no accepted/actionable findings reported.
Observed result after fix: The dev flow runner supports Telegram thinking-final and native sendMessageDraft working-final, both production and fixture progress labels are static, and formatted reasoning filters no longer treat visible Thinking... answers as hidden reasoning.
What was not tested: No live Telegram network send was rerun after the final native/static-label cleanup; earlier local previews drove the UX iteration.

[clawsweeper]

Codex review: needs real behavior proof before merge.

Workflow note: Future ClawSweeper reviews update this same comment in place.

How this review workflow works

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

Summary
This PR adds Telegram channel-message preview flow tooling and changes progress/reasoning preview text to Thinking/Working across Telegram and related channel parsing surfaces.

Reproducibility: yes. for the PR blocker: source inspection shows the PR formatter emits Shelling\n\n..., while the Telegram slow status-reaction test at PR head still matches Shelling\n.... No current-main bug reproduction is applicable because this is a feature/UX PR.

PR rating
Overall: 🧂 unranked krab
Proof: 🦪 silver shellfish
Patch quality: 🧂 unranked krab
Summary: The patch has a focused correctness blocker and only synthetic proof for visible Telegram transport behavior.

Rank-up moves:

• Fix the remaining Telegram progress assertion and rerun the targeted Telegram/channel streaming tests.
• Add redacted live Telegram Desktop or bot-to-bot proof showing Thinking and Working preview flows plus final delivery.
• Update the PR body with proof so ClawSweeper can re-review automatically; if it does not, ask a maintainer to comment @clawsweeper re-review.

What the crustacean ranks mean

• 🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
• 🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
• 🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
• 🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
• 🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
• 🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
• 🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

PR egg
🎁 Pass real behavior proof to wake the egg and unlock a hatchable treat.

Where did the egg go?

• The egg game starts only after the PR passes the real-behavior proof check.
• Before that, no creature, rarity, or ASCII portrait is rolled. The treat waits for real proof.
• This is still just collectible flavor: proof affects review readiness, not creature quality.

Real behavior proof
Needs real behavior proof before merge: The PR body provides targeted tests and autoreview only, and explicitly says no live Telegram network send was rerun after the final parser/timer cleanup. After adding proof, update the PR body; ClawSweeper should re-review automatically. If it does not, the PR author or someone with repository write access can comment @clawsweeper re-review.

Mantis proof suggestion
A native Telegram Desktop proof would directly show preview update behavior, cleanup, and durable final delivery that unit tests cannot prove. A maintainer can ask Mantis to capture proof by posting a new PR comment that starts with the OpenClaw Mantis account mention, followed by:

telegram desktop proof: verify Telegram thinking-final and working-final preview flows show Thinking/Working updates, clear or stop previews correctly, and deliver one durable final answer.

Risk before merge
Why this matters: - No live Telegram network proof after the final parser/timer cleanup means preview clear/stop and durable final delivery remain unproven outside tests.

• Renaming reasoning/progress labels to Thinking/Working changes user-visible wording across channel surfaces, so maintainers should explicitly accept the compatibility impact before merge.
• The PR has a protected maintainer label, so landing needs explicit maintainer handling even after the source blocker is fixed.

Maintainer options:

1. Fix and prove Telegram live (recommended)
   Update the remaining Telegram progress assertion, rerun the targeted tests, and require redacted Telegram proof showing both preview flows update, clear or stop correctly, and deliver one durable final answer.
2. Accept synthetic validation
   Maintainers can choose to merge with fixture/unit validation only if they intentionally accept the live transport timing and delivery uncertainty.
3. Pause the wording change
   If the Thinking/Working terminology is not settled, pause or narrow the PR until the user-visible channel wording policy is approved.

Next step before merge
This needs contributor proof and maintainer/product review; the source defects are concrete, but missing real behavior proof keeps this out of automated repair routing.

Security
Cleared: No concrete security or supply-chain regression was found; the diff does not add dependencies, workflows, permissions, package metadata, or new secret persistence.

Review findings

• [P2] Update the Telegram progress assertion — src/plugin-sdk/channel-streaming.ts:983
• [P3] Fix the Thinking prefix comment — src/config/types.agent-defaults.ts:425

Review details

Best possible solution:

Fix the remaining Telegram progress assertion, rerun the targeted touched-surface tests, and land only after maintainers accept the wording change with redacted live Telegram proof.

Do we have a high-confidence way to reproduce the issue?

Yes for the PR blocker: source inspection shows the PR formatter emits Shelling\n\n..., while the Telegram slow status-reaction test at PR head still matches Shelling\n.... No current-main bug reproduction is applicable because this is a feature/UX PR.

Is this the best way to solve the issue?

No, not yet; the direction is plausible, but the patch needs the remaining Telegram test expectation fixed and live Telegram proof before it is the best mergeable solution.

Label justifications:

• P2: This is a normal-priority user-visible Telegram/channel improvement with a focused source blocker and limited blast radius.
• merge-risk: 🚨 compatibility: The PR changes user-visible progress and reasoning wording from Reasoning/ellipsis labels to Thinking/Working across channel surfaces.
• merge-risk: 🚨 message-delivery: The Telegram preview clear/stop plus durable final delivery path changes and lacks live transport proof after the final cleanup.

Full review comments:

• [P2] Update the Telegram progress assertion — src/plugin-sdk/channel-streaming.ts:983
formatChannelProgressDraftText now inserts a blank line between the label and detail lines, so this flow will update the draft as Shelling\n\n\🛠️ Exec``. The existing Telegram test still matches the old single-newline text, which should fail the targeted test file and leaves this behavior unproven.
  Confidence: 0.93
• [P3] Fix the Thinking prefix comment — src/config/types.agent-defaults.ts:425
  The exported heartbeat config comment now says reasoning messages are prefixed with Thinking., but the new formatter emits Thinking followed by a blank line. Please keep the public type documentation aligned with the actual delivered prefix.
  Confidence: 0.86

Overall correctness: patch is incorrect
Overall confidence: 0.91

Acceptance criteria:

• node scripts/run-vitest.mjs extensions/telegram/src/bot-message-dispatch.test.ts src/plugin-sdk/channel-streaming.test.ts test/scripts/channel-message-flows.test.ts src/infra/heartbeat-runner.returns-default-unset.test.ts src/agents/pi-embedded-utils.test.ts extensions/telegram/src/reasoning-lane-coordinator.test.ts src/plugin-sdk/reply-payload.test.ts src/agents/tools/message-tool.test.ts extensions/discord/src/chunk.test.ts extensions/discord/src/monitor/message-handler.process.test.ts extensions/feishu/src/reply-dispatcher.test.ts

What I checked:

• PR head changes shared progress formatting: At PR head, formatChannelProgressDraftText renders a label followed by a blank line before detail lines, changing the emitted Telegram progress text shape. (src/plugin-sdk/channel-streaming.ts:983, 8941ca0af2a9)
• PR head keeps an old Telegram assertion: The Telegram slow status-reaction test still matches Shelling\n..., while the changed formatter emits Shelling\n\n..., so the targeted test surface is internally inconsistent. (extensions/telegram/src/bot-message-dispatch.test.ts:1828, 8941ca0af2a9)
• Current main behavior baseline: Current main still joins progress label and detail lines with a single newline, which explains why the existing Telegram assertion matched before this PR. (src/plugin-sdk/channel-streaming.ts:977, 6fcfeed5dc67)
• Maintainer Telegram proof standard: The matching Telegram maintainer note says streaming/transport/reply-context changes need real Telegram proof, and this PR body says no live Telegram network send was rerun after the final parser/timer cleanup. (.agents/maintainer-notes/telegram.md:1, 6fcfeed5dc67)
• Feature history routing: Current main blame attributes the relevant progress label/formatter and Telegram progress draft implementation to commit 02f8fb7. (src/plugin-sdk/channel-streaming.ts:99, 02f8fb7147c0)

Likely related people:

• steipete: Git blame on current main attributes the shared progress label/formatter and Telegram progress draft code paths to commit 02f8fb7. (role: recent area contributor; confidence: medium; commits: 02f8fb7147c0; files: src/plugin-sdk/channel-streaming.ts, extensions/telegram/src/bot-message-dispatch.ts)

Codex review notes: model gpt-5.5, reasoning high; reviewed against 6fcfeed5dc67.

[obviyus]

Thanks for the cleanup here. One request before merge: can you wire the label animation into the production Telegram native sendMessageDraft path too, or narrow the PR wording/tests so they do not imply native draft animation ships there?

Right now the dev working-final flow animates native drafts, but production native tool-progress formatting does not pass labelFrame, so sendMessageDraft appears to keep a stable label. Since this PR is UX-facing, the demo flow and production behavior should match.

[Patrick-Erichsen]

Addressed in 8941ca0af2: I took the “narrow the PR wording/tests” path rather than shipping native draft label animation.

Current behavior is now static labels end-to-end: production Telegram native sendMessageDraft tool progress uses Working, the working-final dev flow uses the same static label, and the PR body/proof no longer claims animated ellipses. I also fixed the review follow-ups from autoreview around forum thread handling and the progress gate callback type.

Verification after the cleanup included focused Telegram flow/dispatch tests, cross-channel reasoning/progress tests, and final autoreview clean against origin/main.

---

In plain english, I decided not to whole animated ellipsis thing after our convo about it potentially backing up other edits since Telegram has a 1RPS limit on updates.