docs: adjudicate the intent-sweep anomaly queue (iss-16..iss-23) - #9
Merged
Conversation
The NON-CANONICAL / do-not-implement banner asserted the predecessor's spc-78 spec as implementation-authoritative, inverting the brief's delivery-state provenance doctrine (spc-N attributes, never governs) and leaving itd-65's declared blocker unimplementable. The spc-78 deltas (launch-payload.json override, directory-convention smoke) are now Prior Art design input plus an explicit spec-time Open Question; the body remains canonical per the brief's launch surface. Assisted-by: Claude:claude-fable-5
…ior art The 'AC reconciliation (spc-52 — implementation complete)' section read as a delivery claim while itd-50 sits in planned/. Verified the Go tree carries no audit-loop implementation: the delivery is the predecessor's (spc-52), so the intent stays planned and the table becomes Prior Art design input per the brief's delivery-state provenance note, with the predecessor's open-question answers marked for spec-time re-adjudication. Assisted-by: Claude:claude-fable-5
Phase 0 registers the itd-37 discipline while itd-36 lands in Phase 2 — a phase-order violation only because the frontmatter declared a hard blocked_by edge. The intent's own body pins memory routing as the secondary, gracefully-deferring half and capture + enforcement as independently shippable, so the honest edge is soft: builds_on [itd-1, itd-36], with the split now stated where the fallback is argued. Assisted-by: Claude:claude-fable-5
The phase doc asserted itd-27 depends on itd-42; both intents declare the reverse (itd-42 blocked_by itd-27, itd-27 'Extended by: itd-42', itd-42's press release extends the grill itd-27 built). The dependency bullet and the open-question ordering line now read itd-27 -> itd-42 — the first live catch adjudicated for the itd-78 phase-consistency lint. Assisted-by: Claude:claude-fable-5
…able The intent-sweep flagged itd-65/66/67/72 as a silent stall: critical and major intents in no phase Scope. adr-33 already records the decision (deepenings scheduled separately; Phase 1 owns only the release cut) but the phase index — the sole ownership source — never pointed at it. Beyond Phase 6 now names the deepening cluster explicitly, including itd-67 which rides the same tiering, so the unscheduled state reads as recorded intent, not a stall. Assisted-by: Claude:claude-fable-5
The launch payload manifest, default-deny excludes, and launch.allow opt-in that itd-16's launch-integration bullet leans on are itd-66's render contract. itd-8 already reads as command-surface references with no dangling intent name, so this was the only live occurrence. Assisted-by: Claude:claude-fable-5
The banner claimed itd-6 was superseded outright and preserved as historical record, while Phase 0's Scope schedules it as the oracle adapter seam — under adr-34 a scheduled intent must be planned, so the directory was right and the banner wrong. It now records precisely what ADR-25 superseded (RP as abcd's single integration) and how the body reads today (the RP adapter's contract, adapted at spec time). Assisted-by: Claude:claude-fable-5
…ming The intent still gated on /abcd:launch throughout, which adr-28 makes impossible (the curated release excludes .abcd/** wholesale, so launch never evaluates what the gate guards). Both GWT criteria, the scope bullets, and the test scenario now name the lifeboat gate per the brief's provenance-substrate § 4 vocabulary, and the banner drops its read-as translation rule now that the body says it directly. Assisted-by: Claude:claude-fable-5
Assisted-by: Claude:claude-fable-5
REPPL
added a commit
that referenced
this pull request
Jul 11, 2026
* docs: iss-16 — itd-66 banner demoted to predecessor prior art The NON-CANONICAL / do-not-implement banner asserted the predecessor's spc-78 spec as implementation-authoritative, inverting the brief's delivery-state provenance doctrine (spc-N attributes, never governs) and leaving itd-65's declared blocker unimplementable. The spc-78 deltas (launch-payload.json override, directory-convention smoke) are now Prior Art design input plus an explicit spec-time Open Question; the body remains canonical per the brief's launch surface. Assisted-by: Claude:claude-fable-5 * docs: iss-17 — itd-50 reconciliation table reframed as predecessor prior art The 'AC reconciliation (spc-52 — implementation complete)' section read as a delivery claim while itd-50 sits in planned/. Verified the Go tree carries no audit-loop implementation: the delivery is the predecessor's (spc-52), so the intent stays planned and the table becomes Prior Art design input per the brief's delivery-state provenance note, with the predecessor's open-question answers marked for spec-time re-adjudication. Assisted-by: Claude:claude-fable-5 * docs: iss-18 — itd-37's itd-36 edge downgraded to builds_on Phase 0 registers the itd-37 discipline while itd-36 lands in Phase 2 — a phase-order violation only because the frontmatter declared a hard blocked_by edge. The intent's own body pins memory routing as the secondary, gracefully-deferring half and capture + enforcement as independently shippable, so the honest edge is soft: builds_on [itd-1, itd-36], with the split now stated where the fallback is argued. Assisted-by: Claude:claude-fable-5 * docs: iss-19 — phase-3 itd-27/itd-42 dependency direction corrected The phase doc asserted itd-27 depends on itd-42; both intents declare the reverse (itd-42 blocked_by itd-27, itd-27 'Extended by: itd-42', itd-42's press release extends the grill itd-27 built). The dependency bullet and the open-question ordering line now read itd-27 -> itd-42 — the first live catch adjudicated for the itd-78 phase-consistency lint. Assisted-by: Claude:claude-fable-5 * docs: iss-20 — launch cluster's deliberate unscheduling made discoverable The intent-sweep flagged itd-65/66/67/72 as a silent stall: critical and major intents in no phase Scope. adr-33 already records the decision (deepenings scheduled separately; Phase 1 owns only the release cut) but the phase index — the sole ownership source — never pointed at it. Beyond Phase 6 now names the deepening cluster explicitly, including itd-67 which rides the same tiering, so the unscheduled state reads as recorded intent, not a stall. Assisted-by: Claude:claude-fable-5 * docs: iss-21 — itd-16's itd-launch name reference re-pointed to itd-66 The launch payload manifest, default-deny excludes, and launch.allow opt-in that itd-16's launch-integration bullet leans on are itd-66's render contract. itd-8 already reads as command-surface references with no dangling intent name, so this was the only live occurrence. Assisted-by: Claude:claude-fable-5 * docs: iss-22 — itd-6 banner narrowed to framing-only supersession The banner claimed itd-6 was superseded outright and preserved as historical record, while Phase 0's Scope schedules it as the oracle adapter seam — under adr-34 a scheduled intent must be planned, so the directory was right and the banner wrong. It now records precisely what ADR-25 superseded (RP as abcd's single integration) and how the body reads today (the RP adapter's contract, adapted at spec time). Assisted-by: Claude:claude-fable-5 * docs: iss-23 — itd-36 body reconciled with adr-28's lifeboat gate framing The intent still gated on /abcd:launch throughout, which adr-28 makes impossible (the curated release excludes .abcd/** wholesale, so launch never evaluates what the gate guards). Both GWT criteria, the scope bullets, and the test scenario now name the lifeboat gate per the brief's provenance-substrate § 4 vocabulary, and the banner drops its read-as translation rule now that the body says it directly. Assisted-by: Claude:claude-fable-5 * chore: log adjudication decisions for iss-16..23 drain Assisted-by: Claude:claude-fable-5
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.
Drains the intent-sweep anomaly queue captured in #7: all eight ledger issues
(five majors, three minors) adjudicated and moved to
resolved/, one commitper issue, each with a resolution note recording the reasoning. (Ledger
issues, not GitHub issues — no auto-close keywords.)
Majors
the predecessor's spc-78 spec as implementation-authoritative, inverting the
brief's delivery-state provenance doctrine and leaving itd-65's declared
blocker unimplementable. Demoted to a Prior Art entry; the spc-78 deltas
(config-file payload override, directory-convention smoke) become a
spec-time Open Question; the body stays canonical per the brief's launch
surface.
implementation: the "implementation complete" AC table is the predecessor's
spc-52 delivery. Intent stays
planned/; the table is reframed aspredecessor Prior Art with its open-question answers marked for spec-time
re-adjudication.
typing: itd-37's own design says capture + enforcement ship independently
(the Phase 0 discipline registration) and only extraction-to-memory waits on
itd-36.
blocked_by: [itd-36]→builds_on: [itd-1, itd-36], rationalestated in the body.
themselves (itd-42
blocked_by: [itd-27]; itd-27 "Extended by: itd-42"):the phase-3 doc's reversed dependency bullet and ordering line corrected to
itd-27 → itd-42.
deferral, but the phase index never pointed at it.
phases/README.md§ Beyond Phase 6 now names the deepening cluster (itd-65/66/67/70/72/73,
covering itd-67 which adr-33's list omitted) with the adr-33 pointer and
the itd-78 derived-priority note.
Minors
itd-launchreference re-pointed to itd-66;itd-8 carries no dangling name today.
planned/: Phase 0 schedules it as the oracleadapter seam, and scheduled implies planned (adr-34). The banner is narrowed
from whole-intent supersession to framing-only (ADR-25), with the body read
as the RP adapter's contract.
every launch-gate reference (both GWT criteria, scope bullets, test
scenario) now names the lifeboat restrictive-licence gate run by
/abcd:disembark; the banner drops its read-as translation rule.Checks
make record-lint— zero blockers (remaining WARNs pre-date this branch,none in touched files);
make docs-lint— 0 findings.make preflight— clean (build, gofmt, vet, tests, race)..abcd/work/DECISIONS.md.Assisted-by: Claude