fix(ci): re-pin slack/telegram integration coverage floors to observed reality

[BenKurrek]

What

Re-pins the integration-tier coverage ratchet floors for the two failing crates to the observed numbers from main's own gate output (run 31796056891):

crate	floor (was, 2026-08-04)	observed / re-pinned (2026-08-14)
slack	93.95% · 3697 of 3935	93.57% · 3666 of 3918
telegram	90.31% · 1435 of 1589	56.63% · 2410 of 4256 (post-#7464; see below)

Tolerances, rationale, and issue links are unchanged; only the four captured numbers move per entry.

Why

Every main push since 2026-08-13 12:12 has failed Tests (Reborn) on the Reborn integration-tier coverage report job (first red: run 31698975156; last green main push: 2026-08-12 18:32). The 08-13 merges legitimately shrank both floored crates — code and tests deleted together (slack standard-ops rework era, telegram sticker/voice ingress pruning) — which is exactly the "legitimate shrinkage" case the gate's failure text says to resolve by re-pinning the [[crate]] entry to observed numbers.

Nobody noticed for ~24h because the coverage-report job is planned only on main pushes: the PR lanes and the merge queue both skip it, so the queue kept merging over a red main.

Compatibility / rollback

Data-free CI-config change; revert the commit to restore the old floors. No runtime behavior.

Follow-ups

• The telegram entry must be re-measured when feat(telegram): pair linked devices with the bot channel #7464 lands — happened before merge; the entry is now pinned to main's own post-feat(telegram): pair linked devices with the bot channel #7464 gate output (run 31819915736): 56.63% · 2410 of 4256. The percent drop is denominator growth from the linked-device half, which is deliberately crate-tier covered (the MTProto handshake cannot run in the integration harness); a comment above the entry records the tier split.
• Worth a separate look: whether the coverage-report job should also be planned in merge_group so floor breaches eject in the queue instead of landing on main.

Regression-test exemption: deterministic reproduction is impossible because this PR changes only the coverage gate's own data manifest (four floor numbers in tests/integration/coverage-floor.toml); the consuming gate runs solely on main pushes, so the next main push after merge is the only executable verification, and no unit of production behavior exists to reproduce or pin.

Test Strategy

• Integration: Not applicable — no runtime code changes; the changed file is the gate's own manifest. The gate re-evaluates on the next main push, which is the production verification for this file.
• Crate/unit: Not applicable (CI data file). TOML validated locally with tomllib; entry schema matches the file's documented [[crate]] shape (all four captured fields updated together).
• E2E: Not applicable.

🤖 Generated with Claude Code

https://claude.ai/code/session_01LACC5dKyNy3GNXRNbvgz57

[railway-app]

🚅 Deployed to the ironclaw-pr-7655 environment in ironclaw-ci-preview

Service	Status	Web	Updated (UTC)
ironclaw	✅ Success (View Logs)	Web	Aug 14, 2026 at 5:37 pm

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 91378800-5888-49da-874c-227a02d75658

📥 Commits

Reviewing files that changed from the base of the PR and between 2079d3d and 170cbfd.

📒 Files selected for processing (1)

• tests/integration/coverage-floor.toml

---

📝 Walkthrough

Summary by CodeRabbit

• Tests
  • Updated Slack coverage metadata to 93.57% (3,666 of 3,918 lines).
  • Updated Telegram coverage metadata to 56.63% (2,410 of 4,256 lines).
  • Refreshed coverage capture dates and documented the impact of linked-device code on Telegram integration coverage.

Walkthrough

The integration coverage-floor metadata records updated Slack and Telegram measurements captured on 2026-08-14. The Telegram entry documents coverage dilution from linked-device functionality.

Changes

Coverage floor refresh

Layer / File(s)	Summary
Coverage floor metadata tests/integration/coverage-floor.toml	The Slack floor changed to 93.57% with 3,666 of 3,918 lines. The Telegram floor changed to 56.63% with 2,410 of 4,256 lines. The Telegram entry documents linked-device coverage dilution.

Estimated code review effort: 1 (Trivial) | ~2 minutes

Merge Risk: 🔴 Critical · up to 170cb

Although this PR only changes CI coverage-floor data, the current manifest is reported to contain an extra crate header that can make the coverage gate fail; merge should be blocked until the telegram entry structure is corrected.

Suggested reviewers: serrrfirat

🚥 Pre-merge checks | ✅ 3 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description explains the change and validation, but it omits most required template sections, including linked issue, security, database, blast radius, and review track.	Complete the repository template, or explicitly mark non-applicable sections and set the review track to C for the CI change.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title uses Conventional Commits style and accurately describes the coverage-floor metadata change.
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.

✨ Finishing Touches 💡 1

🛠️ Fix failing CI checks 💡

• Create stacked PR
• Commit on current branch

---

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.}

[ironloopai]

🧭 IronLoop Run · Review

This comment updates in place as the Run moves through its stages.

🟩 Final result · Completed

🟨 Queued → 🟦 Working → 🟦 Posting results → 🟩 Completed

_{Automatic trigger · attempt 1 of 3 · completed in 1m}

IronLoop completed the review and posted it to GitHub.

🔗 Result

Open submitted review →

Run details

Run: 6c30a1c9-bc18-45bd-9a7a-b595255b3d7f
Base: main at 7b96a76
Head: fix/integration-coverage-floors-0814 at 2079d3d
Created: 2026-08-14 13:15 UTC
Updated: 2026-08-14 13:16 UTC

[ironloopai]

🔍 IronLoop review

🟢 No actionable findings

No actionable issues found in the coverage-floor re-pin.

Validation

• ✅ Coverage floor manifest — The updated Slack and Telegram line counts parse successfully, and each percentage matches its captured covered/total-line values when rounded to two decimals.
• ✅ Pull request coverage gate — The captured integration-tier coverage report completed successfully with the updated manifest.

Review details

• Run: 6c30a1c9-bc18-45bd-9a7a-b595255b3d7f
• Workflow: Review
• Attempts: 1