WeKnora/internal/agent
lyingbug 62f7da5eea
feat(agent): persist compaction checkpoints across turns (#3407)
* feat(agent): persist compaction checkpoints across turns

Agent compaction only lived inside the turn that ran it: the summary was
discarded at turn end and LoadAgentHistory rebuilt raw history from the
messages table, so every later turn past the threshold paid for another
summarization of nearly the same history.

A compaction whose history summary ends exactly on a stored turn is now
written onto that turn's assistant message (messages.context_checkpoint).
LoadAgentHistory starts from the newest checkpoint: the summary replaces
that turn and everything before it, and later turns are replayed verbatim.
The compactor already treats a leading summary as the previous summary, so
checkpoints fold forward incrementally.

History messages carry their turn's assistant message ID (chat.Message
TurnID, engine-internal) so compaction can tell whether its cut lands on a
stored-turn boundary. Nothing is persisted when the cut ends inside a stored
turn, covers only the live turn, or the history summary degraded to a raw
archive; a split turn's prefix summary is never persisted because that turn
is replayed whole next time. Writes are column-scoped and best-effort, and a
failed lookup falls back to history without a checkpoint.

Storing the checkpoint on the covered turn keeps the boundary implicit, so
session forks copy it with the turn and deleting the turn drops it. The
HistoryTurns cap still applies, now to the turns after the checkpoint.

* test(skill): wait for the background install in the name-conflict test

InstallSkill queues the install on a goroutine, and the fixture row already
starts at installing, so reading the status right after the call raced the
mock install and failed whenever it landed on ready first. Wait for the run
to finish on the reused row instead, as the other install tests do.

* feat(agent): size agent history by tokens instead of turn count

Agent history was capped at history_turns (default 5) because every turn
used to rebuild raw history and compaction could not persist. With
compaction checkpoints persisted, the cap is no longer what keeps a session
inside the window, and it threw away context that would have fit.

LoadAgentHistory now loads the newest checkpoint summary plus the newest
turns after it that fit a token budget: the run's compaction threshold
(agent.HistoryTokenBudget, the window less the reply reserve). History
that large already triggers the first round's compaction, which persists a
new checkpoint, so compaction rather than a turn count bounds the session.
Loading more would only hand that compaction a summarizer input larger
than the model can read.

Rows are paged backwards by (created_at, id) and the read stops at the
checkpoint or once the budget is full, so a long session is never read
whole. Kept turns stay contiguous, the newest turn is kept even when it
alone exceeds the budget (compaction splits it), and a turn cut by the read
is left out rather than replayed without its original question.

Agent mode no longer reads history_turns: AgentConfig.HistoryTurns is
removed, the editor hides the input in smart-reasoning mode and explains
the automatic behavior, and the smart-reasoning builtins drop the setting.
Quick-answer (KnowledgeQA) keeps its turn cap since it has no compaction.

* Revert "feat(agent): size agent history by tokens instead of turn count"

This reverts commit d05fb6e5f1f161813a88e9127609f7a059f38f12.

Loading history up to the compaction threshold lets the loader drop the
oldest turns before compaction can trigger: whenever a single turn's
replay is larger than the system prompt plus the new question, the
trimmed history stays under the threshold, nothing is summarized, and the
session degrades into a large sliding window that never gets a
checkpoint. A turn-by-turn simulation of LoadAgentHistory with the real
compactor showed zero compactions over 60 turns in those cases.

Keep the persisted checkpoints with the HistoryTurns cap for now; token
budgeted history returns in a separate change with a budget that leaves
the overflow to compaction.
2026-09-18 20:38:56 +08:00
..
approval refactor(agent): discover and invoke MCP tools on demand 2026-09-08 12:01:36 +08:00
compaction feat(agent): persist compaction checkpoints across turns (#3407) 2026-09-18 20:38:56 +08:00
skills fix(skills): verify runtime prerequisites and support installer steering 2026-09-09 23:04:59 +08:00
token feat(agent): keep long sandbox turns from truncating, overflowing, or rewriting files 2026-09-02 22:09:09 +08:00
tools fix(agent): allow keyword retry after search_knowledge rerank rejection 2026-09-18 16:06:52 +08:00
act.go refactor(agent): consolidate knowledge retrieval tools into search_knowledge, read_document, list_documents (#3381) 2026-09-18 14:12:22 +08:00
act_order_test.go refactor(agent): consolidate knowledge retrieval tools into search_knowledge, read_document, list_documents (#3381) 2026-09-18 14:12:22 +08:00
checkpoint.go feat(agent): persist compaction checkpoints across turns (#3407) 2026-09-18 20:38:56 +08:00
const.go fix(browser): support human handoffs and resume interrupted tasks 2026-09-11 23:56:13 +08:00
const_test.go fix(browser): support human handoffs and resume interrupted tasks 2026-09-11 23:56:13 +08:00
context_debug.go feat(agent): keep long sandbox turns from truncating, overflowing, or rewriting files 2026-09-02 22:09:09 +08:00
engine.go feat(agent): persist compaction checkpoints across turns (#3407) 2026-09-18 20:38:56 +08:00
engine_image_test.go feat: auto-describe MCP tool result images via VLM 2026-03-25 21:38:15 +08:00
engine_memory_test.go feat(memory): add cross-session long-term memory 2026-08-13 18:00:02 +08:00
engine_test.go feat(agent): persist compaction checkpoints across turns (#3407) 2026-09-18 20:38:56 +08:00
finalize.go feat(browser): polish connection settings, open tasks in a new window, and streamline prompts (#3193) 2026-09-11 21:28:58 +08:00
grounding_prompt.go fix(agent): search bound knowledge bases before answering, and pass suggested-question origins (#3400) 2026-09-18 17:00:47 +08:00
grounding_prompt_test.go fix(agent): search bound knowledge bases before answering, and pass suggested-question origins (#3400) 2026-09-18 17:00:47 +08:00
knowledge_base_profile_prompt_test.go feat(knowledge): derive knowledge base descriptions from document profiles (#3370) 2026-09-17 22:21:55 +08:00
langfuse_test.go refactor(agent): introduce modelcontext registry and fix handle-codec bugs 2026-07-27 11:05:43 +08:00
mcp_exposure_test.go fix(mcp): normalize JSON-encoded argument envelopes 2026-09-09 16:11:52 +08:00
mcp_proxy_test.go fix(mcp): normalize JSON-encoded argument envelopes 2026-09-09 16:11:52 +08:00
observe.go feat(agent): persist compaction checkpoints across turns (#3407) 2026-09-18 20:38:56 +08:00
observe_test.go fix(agent): search bound knowledge bases before answering, and pass suggested-question origins (#3400) 2026-09-18 17:00:47 +08:00
prompt_composition_test.go fix(agent): search bound knowledge bases before answering, and pass suggested-question origins (#3400) 2026-09-18 17:00:47 +08:00
prompts.go fix(agent): search bound knowledge bases before answering, and pass suggested-question origins (#3400) 2026-09-18 17:00:47 +08:00
prompts_browser.go fix(browser): align BrowserSkill 0.3.0 and repair task lifecycle 2026-09-17 11:54:12 +08:00
prompts_browser_test.go feat(browser): polish connection settings, open tasks in a new window, and streamline prompts (#3193) 2026-09-11 21:28:58 +08:00
prompts_shell_test.go fix(sandbox): separate file access from artifact publishing (#3356) 2026-09-17 16:18:01 +08:00
prompts_wiki.go Remove redundant Wiki operation log 2026-07-28 11:18:33 +08:00
prompts_wiki_test.go refactor(agent): introduce modelcontext registry and fix handle-codec bugs 2026-07-27 11:05:43 +08:00
steer.go fix(chat): preserve steering continuity and composer focus 2026-09-09 20:15:23 +08:00
steer_test.go fix(chat): preserve steering continuity and composer focus 2026-09-09 20:15:23 +08:00
terminal_event_test.go fix(agent): stop emitting terminal answer events for retryable empty stops 2026-09-14 21:02:15 +08:00
think.go refactor(modelcontext): consolidate source_refs parsing, fixed data_analysis table, and identifier leak report (#3372) 2026-09-17 22:08:16 +08:00
tool_images.go fix(browser): complete task cancellation, screenshots, and result display (#3206) 2026-09-12 09:32:26 +08:00
tool_images_test.go fix(browser): complete task cancellation, screenshots, and result display (#3206) 2026-09-12 09:32:26 +08:00
turn_usage_test.go feat(agent): report and attribute per-turn LLM token usage 2026-08-24 10:36:19 +08:00