docs(handoff): two of this doc's own open blocks were answered weeks ago and still read as live - #1476
Merged
Merged
Conversation
…ago and still read as live Close-out sweep after the yield arc merged (#1399). Found by asking what the arc left behind rather than by any check — nothing prompts this. * `### A repo that MEASURES but whose every doc is unreadable …` and `### The same hazard has now appeared in FOUR spellings …` were both CLOSED by #1267. Both kept a present-tense heading and a `Next probe` under a section titled `Open investigations — live diagnosis state`, with their own `RESOLVED —` counterparts 16 and 55 lines below them. A /resume following step 3 would have re-derived work that shipped weeks ago — the exact hazard this doc's own gotcha names, left standing in the doc that records it. Re-headed as SUPERSEDED with pointers; bodies kept for their values. * 🔴 A RESOLVED BLOCK DOES NOT RETIRE THE OPEN ONE. That is a separate edit and nothing prompts it. Recorded as a gotcha with the check that would have caught it (`grep -n '^### '` over the file). Four /resume runs read this doc during the arc, mine included, and none caught it. * Filed rank 3: `ship.sh` cannot reach the laptop off-LAN. `host-role.sh` defines LAPTOP_IP_SECONDARY (nebula) beside the primary, but the SSH default derives from the primary only, so the remote leg dies rc 255 from off-network. Measured shipping 4a67ea7 — the laptop converged only under a manual REMOTE_SSH. It fails honestly, but it splits the run in two, and ship.sh then prints `cross-host agreement NOT COMPARED` — which is the check that exists to catch the hosts landing on different commits. Closing condition given. Also closed the cairn `devrc/handoff-index` entry's OPEN bullet, which had asked for exactly the yield answer this arc produced. Appending a RESOLVED bullet did NOT retire it (the badge still read `🔴 1 OPEN`), and the first closure named no sha so it scored `⚠ 1 UNVERIFIABLE` — both fixed in place; the entry now shows no badges and 26/26 parse. Content gates run against this diff (it adds two private-IP literals): test_no_public_ips, test_no_client_hostnames, test_no_captured_text, test_no_captured_markup — 227 passed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SwwnnDzefhhE8cG9sqjafE Claude-Session-Id: f0fffa8e-a2ab-464e-bd7c-dfe9bd8955f0
…ndex-arc-closeout Claude-Session-Id: f0fffa8e-a2ab-464e-bd7c-dfe9bd8955f0
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Close-out sweep after the yield arc merged (#1399), prompted by asking what the arc left behind. Nothing checks for this.
The finding
claudedocs/handoff-handoff-search-index.mdcarried two###blocks underOpen investigations — live diagnosis statethat were closed by #1267. Both kept a present-tense heading and aNext probe, with their ownRESOLVED —counterparts 16 and 55 lines below them in the same section.A
/resumefollowing step 3 reads that section as live diagnosis state. It would have re-derived work that shipped weeks ago — which is exactly the hazard this document's own gotcha names:Left standing in the document that records it. Four
/resumeruns read this doc during the arc, mine included, and none caught it.Why it survived
🔴 A
RESOLVEDblock does not retire theOPENone. Writing the resolution and retiring the open block are two separate edits, and only the first one feels like progress. Recorded as a gotcha, with the check that would have caught it in one command:grep -n '^### 'over the file.Both blocks are re-headed
SUPERSEDEDwith pointers to their resolutions; the bodies stay, because their measured values are still worth reading.Also in this diff
Filed as rank 3 —
ship.shcannot reach the laptop off-LAN.host-role.shdefinesLAPTOP_IP_SECONDARY(nebula) beside the primary, but the SSH default derives from the primary only, so from off-network the remote leg diesConnection timed out, rc 255. Measured shipping4a67ea73: the laptop converged only under a manualREMOTE_SSH.It fails honestly rather than silently — but it splits the run in two, and
ship.shthen printscross-host agreement NOT COMPARED, which is the check that exists to catch the two hosts landing on different commits. Closing condition is stated in the item.Outside this diff
The cairn
devrc/handoff-indexentry'sOPEN:bullet asked for exactly the yield answer this arc produced. Closing it demonstrated the same lesson twice:RESOLVED:bullet did not retire theOPEN:one — the badge still read🔴 1 OPEN;⚠ 1 UNVERIFIABLE.Both fixed in place. The entry now shows no badges, names a sha that resolves, and 26/26 entries parse.
Gate
Docs-only, but the diff adds two private-IP literals, so the four content gates were run against it:
test_no_public_ips,test_no_client_hostnames,test_no_captured_text,test_no_captured_markup— 227 passed.🤖 Generated with Claude Code
https://claude.ai/code/session_01SwwnnDzefhhE8cG9sqjafE