amadan
🤖 spec: cite the ref, not the claim
The suite session's lesson and the better of the two. A claim from a peer gets less scrutiny than a claim from a file and should get more: a file can be opened at the commit you name, a peer's claim carries no provenance at all — you cannot see which checkout they read or whether they fetched. Both of this week's cross-repository mistakes have that shape. A doctrine ruling made on something memoria told the suite, and a memoria merge made on something the suite implied. Also withdrawn on the record: the S1 low-water mark I proposed for the restore signal. It requires ordering revision tokens, which §4.1 forbids four sections away in the paragraph explaining why STALE and AHEAD are well-defined — they are the sibling's verdict on its own log precisely so that memoria never compares two revisions. And a low-water mark describes pruning rather than restore: pruning raises the earliest, a restore lowers the latest. What survives is better. A whole-log restore rewinds the cursor and S1a re-announces; a PARTIAL restore rewinds item state without moving the feed position, so nothing fires and only a per-item answer can see it. That is exactly AHEAD, and it makes the two mechanisms complementary rather than overlapping.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
1 file changed,
+12
−0
docs/superpowers/specs/2026-09-03-memoria-design.md+12 −0
diff --git a/docs/superpowers/specs/2026-09-03-memoria-design.md b/docs/superpowers/specs/2026-09-03-memoria-design.md| index 98c1d84..6ee04a1 100644 |
| --- a/docs/superpowers/specs/2026-09-03-memoria-design.md |
| +++ b/docs/superpowers/specs/2026-09-03-memoria-design.md |
| @@ -574,6 +574,18 @@ read a ref you did not just write.** It is in suite AGENTS.md already, |
| which is the point — a rule nobody runs is a rule that catches nothing, |
| and it is now three-for-three at catching nothing. |
| +**And a second lesson from the same exchange, which is the suite session's |
| +and is the more useful of the two: cite the ref, not the claim.** A claim |
| +arriving from a peer gets *less* scrutiny than one arriving from a file, |
| +and it should get more — a file can be opened at the commit you name, |
| +while a peer's claim carries no provenance at all. You cannot see which |
| +checkout they read, whether they fetched, or whether they were quoting or |
| +remembering. Both of this week's cross-repository mistakes have that |
| +shape: a doctrine ruling made on something memoria told the suite, and a |
| +memoria merge made on something the suite implied. **When you tell another |
| +repository a fact about this one, give the commit it is true at. When one |
| +tells you, check it before you rule on it.** |
| + |
| **One finding was wrong, and how it was wrong is worth as much.** The |
| review could not resolve `6797bbf` or `4aea8f4` and said §13's blocker was |
| therefore not shown closed. The hashes are correct — they are |