diff --git a/.abcd/development/intents/drafts/itd-6-terminal-capture.md b/.abcd/development/intents/superseded/itd-6-terminal-capture.md
similarity index 84%
rename from .abcd/development/intents/drafts/itd-6-terminal-capture.md
rename to .abcd/development/intents/superseded/itd-6-terminal-capture.md
index 63f8ba5..cc1c214 100644
--- a/.abcd/development/intents/drafts/itd-6-terminal-capture.md
+++ b/.abcd/development/intents/superseded/itd-6-terminal-capture.md
@@ -7,8 +7,11 @@ suggested_kind: null
 reclassification_history: []
 builds_on: []
 severity: major
+superseded_by: itd-11
 ---
 
+> **Superseded by itd-11 (terminal-cast-import), shipped 2026-09-12.** The evidence this intent asked for — a terminal session's commands and output on the session clock, joined to narration by `merge` and rendered by `report` — is delivered by `testimony import`, which takes an operator-recorded asciicast (v2 or v3) after the session instead of wrapping a recorder inside `record`. The three open questions below are answered there: the cast header's whole-second timestamp anchors the clock (with `-offset` as the correction), output is kept verbatim and split at the JSONL budget rather than truncated, and TUI redraw traffic is preserved in the archived cast but not coalesced. Kept as the historical record of the wrap-the-shell design.
+
 # Think Aloud While You Work at the Terminal
 
 ## Press Release
diff --git a/.abcd/work/DECISIONS.md b/.abcd/work/DECISIONS.md
index cdc2139..9648359 100644
--- a/.abcd/work/DECISIONS.md
+++ b/.abcd/work/DECISIONS.md
@@ -1676,3 +1676,4 @@ Architecture-shaping decisions graduate to an ADR under
   CSI residue, as designed and documented — logged as `iss-2609120520334220`.
   The 3.x (v3) recorder is not installed, so v3 stays fixture-verified only.
 - 2026-09-15 — PR bodies carry the `Assisted-by:` trailer as their last line, matching the global attribution gate that refuses a body without it; the earlier "no footer in PR bodies" rule contradicted that gate and is replaced. Session links stay out of PR bodies and commits.
+- 2026-09-15 — itd-6 (terminal capture inside `record`) is superseded by itd-11: `import` delivers the same evidence from an operator-recorded cast without touching `record`'s lifecycle. Moved to `intents/superseded/` with a `superseded_by` pointer; abcd has no supersede verb, so the move is by hand (captured as a tooling observation).
diff --git a/.abcd/work/issues/.iss-alloc.lock b/.abcd/work/issues/.iss-alloc.lock
new file mode 100644
index 0000000..e69de29
diff --git a/.abcd/work/issues/open/iss-2609150752529778-abcd-intent-has-no-verb-to-retire-an-intent-as-superseded-th.md b/.abcd/work/issues/open/iss-2609150752529778-abcd-intent-has-no-verb-to-retire-an-intent-as-superseded-th.md
new file mode 100644
index 0000000..e10bd66
--- /dev/null
+++ b/.abcd/work/issues/open/iss-2609150752529778-abcd-intent-has-no-verb-to-retire-an-intent-as-superseded-th.md
@@ -0,0 +1,14 @@
+---
+schema_version: 1
+id: "iss-2609150752529778"
+slug: "abcd-intent-has-no-verb-to-retire-an-intent-as-superseded-th"
+severity: "minor"
+category: "observation"
+source: "user-observation"
+found_during: "itd-6-supersession"
+origin: researcher-authored
+production_mode: hand-written
+found_at: ".abcd/development/intents/README.md"
+---
+
+abcd intent has no verb to retire an intent as superseded: the lifecycle table documents intents/superseded/ but the only way to move a draft there is git mv plus a hand-typed superseded_by frontmatter key, so the record-id seam never sees the transition