docs(evidence): fresh deploy lag — merged PR #43 AI-search evidence is on main but not live (past 5864e39) - #153
Conversation
…udit still teaches the retired Agent Desk Merged PR #43 (AI-search evidence re-run + q5 truth aligned to the current offer) is on origin/main ad9cee3 but not live: release-state pins 5864e39 (deployed before the merge), so live /audit still embeds the 2026-08-06 artifact with the retired Agent Desk framing, contradicting live llms.txt/offer.md. Receipt pins the gap (md5s, testedOn, q5 truth), verifies all other pages match main, names the deployment blocker (expired wrangler auth, Nish-reserved, already NOTIFY'd), and confirms npm run check/test pass on the to-be-shipped revision.
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
nish3451 has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.
📝 WalkthroughWalkthroughThe PR adds an evidence receipt for a stale ChangesDeployment lag evidence
Estimated code review effort: 1 (Trivial) | ~3 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🧹 Nitpick comments (1)
docs/evidence/ai-search-evidence-lag-2026-08-12.md (1)
34-38: 🗄️ Data Integrity & Integration | 🔵 Trivial | ⚡ Quick winMake the acceptance target reproducible.
origin/main ≥ ad9cee3permits later commits, but the recorded/auditvalues were measured atad9cee3. If a later commit changespublic/audit.html, this receipt no longer validates that deployment. Pin the closeout toad9cee3, or rerun the comparisons for the actual deployed SHA.Also applies to: 109-110
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/evidence/ai-search-evidence-lag-2026-08-12.md` around lines 34 - 38, Update the deployment acceptance criteria in the evidence document to pin fleet-release/wrangler deployment and `/audit` artifact validation to the exact commit `ad9cee3`, rather than allowing later `origin/main` commits; alternatively, regenerate the recorded comparisons and values for the actually deployed SHA. Apply the same correction to the related acceptance entries.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/evidence/ai-search-evidence-lag-2026-08-12.md`:
- Around line 60-61: Update the release timestamp in the evidence entry for sha
5864e39274c830440ebc3dbd0a26f03865c8961d to include its timezone offset, using
the same timezone basis as the PR `#43` merge timestamp so the stated interval is
reproducible.
- Around line 85-90: Revise the status sentence in the evidence document to
clearly separate the statuses: identify `#140` as the open ship-verify PR, then
separately identify `#141` as merged, while preserving the existing SHA and
citation details.
- Around line 63-70: Update the evidence section’s artifact comparison table to
replace MD5 hashes with SHA-256 digests, or record reproducible direct
byte-comparison output for each URL/source pair. Add comparison results for
every listed page, including the additional entries referenced by the review,
while preserving the pinned-release and origin/main comparisons.
---
Nitpick comments:
In `@docs/evidence/ai-search-evidence-lag-2026-08-12.md`:
- Around line 34-38: Update the deployment acceptance criteria in the evidence
document to pin fleet-release/wrangler deployment and `/audit` artifact
validation to the exact commit `ad9cee3`, rather than allowing later
`origin/main` commits; alternatively, regenerate the recorded comparisons and
values for the actually deployed SHA. Apply the same correction to the related
acceptance entries.
🪄 Autofix
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: fa70f1da-209a-4f93-a972-ba5eae9bd28d
📒 Files selected for processing (1)
docs/evidence/ai-search-evidence-lag-2026-08-12.md
| `sha 5864e39274c830440ebc3dbd0a26f03865c8961d` (2026-08-12T06:37:42) — | ||
| deployed ~1h47m **before** PR #43 merged (08:24:14+05:30). |
There was a problem hiding this comment.
🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick win
Add the timezone to the release timestamp.
2026-08-12T06:37:42 has no timezone, while the merge timestamp uses +05:30. The stated ~1h47m interval is not reproducible until both timestamps use the same timezone.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@docs/evidence/ai-search-evidence-lag-2026-08-12.md` around lines 60 - 61,
Update the release timestamp in the evidence entry for sha
5864e39274c830440ebc3dbd0a26f03865c8961d to include its timezone offset, using
the same timezone basis as the PR `#43` merge timestamp so the stated interval is
reproducible.
| 2. Live `/audit` is byte-identical to the pinned release and still teaches | ||
| the retired Agent Desk: | ||
|
|
||
| | Check | Result | | ||
| |---|---| | ||
| | md5 of live `https://tinystudio.io/audit` | `4018721f9f657c1400a5b38048e9a7c3` | | ||
| | md5 of `public/audit.html` at `5864e39` | `4018721f9f657c1400a5b38048e9a7c3` (identical) | | ||
| | md5 of `public/audit.html` at `origin/main` (`ad9cee3`) | `72116e4a6daa6a26280b647e8ad171be` | |
There was a problem hiding this comment.
🔒 Security & Privacy | 🟠 Major | ⚡ Quick win
Use reproducible, collision-resistant artifact comparisons.
The receipt uses MD5 to support “byte-identical” claims. MD5 is not collision-resistant. The other listed pages also have no recorded digest or direct comparison result. Use SHA-256 or record direct byte-comparison output for every URL/source pair.
Also applies to: 78-83
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@docs/evidence/ai-search-evidence-lag-2026-08-12.md` around lines 63 - 70,
Update the evidence section’s artifact comparison table to replace MD5 hashes
with SHA-256 digests, or record reproducible direct byte-comparison output for
each URL/source pair. Add comparison results for every listed page, including
the additional entries referenced by the review, while preserving the
pinned-release and origin/main comparisons.
| 3. No open PR carries this ship. Open ship-verify PRs #140 (past `b004c11`, | ||
| structured data) and the merged #141 (past `fa8d83c`, App Store citation) | ||
| cover older SHAs; #139 covers the `#28/#30` fleet-release item at | ||
| `18128e8`; #146 closes the PR #42+#43 serial-merge review item (both | ||
| merged) without verifying live deployment of #43's content. Nothing open | ||
| mentions `5864e39` or `ad9cee3`. |
There was a problem hiding this comment.
🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
Separate the open and merged PR statuses.
“Open ship-verify PRs #140 ... and the merged #141” is ambiguous and can make PR #141 appear open. Use “Open ship-verify PR #140 ...; merged PR #141 ...” instead.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@docs/evidence/ai-search-evidence-lag-2026-08-12.md` around lines 85 - 90,
Revise the status sentence in the evidence document to clearly separate the
statuses: identify `#140` as the open ship-verify PR, then separately identify
`#141` as merged, while preserving the existing SHA and citation details.
There was a problem hiding this comment.
nish3451 has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.
1 participant
Closes the fresh deploy-lag class finding (same class as the closed ship items 676/703/716/742/934/953/994): merged PR #43 (
fix/ai-search-rerun-entity-offer, evidence re-run + q5 truth aligned to the current offer) is on origin/mainad9cee3but not live.What was verified (2026-08-12)
8606b0c+ed62202are ancestors of origin/mainad9cee3and are the onlypublic//src/commits in5864e39..origin/main.release-state-tinystudio-io.jsonpins5864e39(deployed 2026-08-12T06:37:42) — ~1h47m before PR evidence(ai-search): re-run controlled entity-and-offer questions after the llms/offer mirror went live #43 merged (08:24:14+05:30)./auditis byte-identical to the pinned release (md54018721f...both) and still embedstestedOn: 2026-08-06with q5 truth "the leak audit, plus the Agent Desk behind it" — the retired product framing — while origin/main carriestestedOn: 2026-08-09and the human-reviewed-desk truth. Livellms.txt/offer.mdalready teach the current offer, so live/auditcontradicts them./,/agents,/pricing,/specimen,/brief-requested,/agent-desk,llms.txt,offer.mdall byte-identical to their sources./auditis the only divergence.5864e39orad9cee3.wrangler is not authenticated (login expired?), NOTIFY'd 2026-08-12T08:54:07; localwrangler whoamiconfirms. Restoring the wrangler login /CLOUDFLARE_API_TOKENis the only remaining action; the next fleet-release tick then shipsad9cee3and this item's acceptance becomes verifiable live.npm run checkpasses andnpm testpasses — 92 tests, 0 failures (headings 6, sitemap 7, worker 55, ui 16, contract 8) onad9cee3. Browser walk (desktop + mobile 390x844, all seven pages): no overflow, no console errors beyond the owned/brief-requestedgtag-CSP item (PR fix(worker): make the Google Ads conversion tag env-driven instead of a dead placeholder #136).Files
docs/evidence/ai-search-evidence-lag-2026-08-12.md— receipt pinning the gap (md5s, testedOn, q5 truth), the blocker, and the acceptance criteria for re-verification once credentials are restored.No product code changed; the fix is merged and merely undelivered.
Summary by CodeRabbit