Skip to content

Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors

Rust

docs: clarify kimi handoff linking and record adapter limitations

Lucas Oliveira96441fa6af09
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 41b640d9..25165fd4 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -26,7 +26,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
`packages/agent-core/src/session/index.ts`), so the previous hook consumed
pending handoffs — legacy and managed — without ever showing them to the
model. `UserPromptSubmit` stdout is injected as a user message before the
- turn. Re-run `ai-memory install-hooks --agent kimi-code --apply` after
+ turn. The native `ai-memory hook` path now also accepts the
+ `user-prompt-submit` event token alongside `user-prompt` for kimi handoff
+ delivery. Re-run `ai-memory install-hooks --agent kimi-code --apply` after
upgrading.
## [1.17.3] - 2026-07-22
diff --git a/docs/managed-workstreams.md b/docs/managed-workstreams.md
index 41c9b53c..ab519ae5 100644
--- a/docs/managed-workstreams.md
+++ b/docs/managed-workstreams.md
@@ -87,10 +87,13 @@ consume the later adoption opportunity.
actual native session and injects only the portable events that session has
not seen. Crush, which has no SessionStart hook, receives the same bounded
packet through a temporary `options.global_context_paths` entry. Kimi Code
- fires SessionStart but discards its stdout, so the kimi adapter links there
- and delivers the packet through the UserPromptSubmit hook instead, whose
- stdout Kimi Code injects as a user message before the turn. Direct
- launches continue to use the existing single-use handoff path.
+ fires SessionStart but discards its stdout, so the kimi adapter's
+ SessionStart hook only captures the event — it neither fetches nor links.
+ The UserPromptSubmit hook issues the `/handoff` GET with the native
+ `session_id` in the query; the server links the session and renders the
+ packet atomically, and Kimi Code injects the hook's stdout as a user
+ message before the turn. Direct launches continue to use the existing
+ single-use handoff path.
4. When the child exits, ai-memory reads the native transcript store without
modifying it. Visible user/assistant messages, completed tool calls/results,
compaction summaries, and a non-mutating Git checkpoint enter an append-only
@@ -181,7 +184,13 @@ Known Kimi Code adapter limitations: subagent transcripts
(`agents/<id>/wire.jsonl` other than `main`) are not imported in v1 and are
recorded as an extraction-loss annotation; the session bucket directory name
is a one-way hash of the working directory, so discovery always reads
-`state.json`'s `workDir` and never parses the bucket name. The native
+`state.json`'s `workDir` and never parses the bucket name. Event ids derive
+from the SHA-256 of the raw wire.jsonl line, so two byte-identical lines —
+only possible with identical content in the same millisecond, because Kimi
+Code stamps each record with `time` — collapse into a single ledger event.
+Legacy sessions that keep `wire.jsonl` directly in the session directory
+(the pre-`agents/` layout the kimi session-store still reads through its
+stat fallback) are neither discovered nor imported in v1. The native
contract was verified against Kimi Code v0.28.1.
Crush needs no ai-memory hook installation for managed mode. The launcher reads