Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors
Rust
Merge remote-tracking branch 'origin/release/3.3' into feat/1645-lossless-project-merge
AkitaOnRails5ad3729d09b0
diff --cc CHANGELOG.mdindex 1964c124,e79a3f65..d5adce61--- a/CHANGELOG.md+++ b/CHANGELOG.md@@@ -13,22 -13,17 +13,33 @@@ version in [`docs/changelog/`](docs/cha### Added+- `ai-memory merge-project --workspace <w> --from <source> --into <target>`+ losslessly joins two project identities inside one workspace. Its default is+ an exact dry run; `--confirm` re-stamps page history, sessions, observations,+ handoffs, embeddings and managed workstreams in one writer transaction, then+ removes the source project. Same-path pages get a new target version that+ appends the source body behind an `ai-memory:merged-from` marker while both+ prior histories remain restorable. The target frontmatter wins, a one-sided+ expiry is removed, and the combined page is flagged stale for consolidation.+ `restore-page --page-id <uuid>` restores either retained history even after+ the source identity is removed. Open sessions and live managed runs refuse+ the merge (#1645).+- `audit-contamination` now reports projects in one workspace whose names+ collide after Unicode compatibility decomposition, case folding, and+ removal of non-spacing marks. Findings distinguish case from accent+ differences and include page/session counts plus last activity (#1645).++ - Added an offline `ai-memory-eval abstention` benchmark with versioned+ clean-room answerable, hard-negative, conflicting, partial and+ missing-channel fixtures; raw lexical/score/density/belief signals;+ FSR/FAR/coverage threshold sweeps with undefined denominators preserved;+ optional reuse of pinned LongMemEval `*_abs` cases; and optional structured+ LLM answerability through the existing QA provider path. Conflicting and+ partial cases remain separate from binary ground truth, no threshold is+ presented as calibrated, and production runtime behavior is unchanged.+ The retrieval evaluator now also parses production's structured+ `{text,citations}` answer shape instead of a stale string (#1717).+- Added `[briefing] inject_on_subagent_start` to `.ai-memory.toml`. A projectthat sets it hands each `claude-code` subagent the project brief when thesubagent starts: the same pinned, `_rules/` and `_slots/` brief that@@@ -41,14 -36,14 +52,6 @@@`hooks/claude-code` shell and PowerShell scripts deliver it, and every markerparser treats the key as a settings boundary. It is off by default (#1715).--- `ai-memory audit-contamination` reports projects whose names differ only by-- spelling: two rows whose names are equal after case folding, Unicode-- composition and separators (`renato` / `Renato` / `Rénato`). That is the-- signature of one working copy reaching the server twice, which no-- cwd-anchored check can see; each finding names the peer projects and their-- page and session counts, and suggests a marker or a merge. Read-only, and it-- never fires for a project that merely shares a prefix (#1645).--- Added an optional `[query_answer] max_context_tokens` admission budget foranswer synthesis. It preserves retrieval hits, reports omitted or clippedevidence, prioritizes known newer corrections, prefers whole snippets and