Commit Graph
9 Commits
Author SHA1 Message Date
Tianyi Cui e8eddc7ef8 Rename RFCs to Agent Notes 2026-07-19 22:52:03 +08:00
Tianyi Cui a278374e8f docs: structure model experience fields 2026-07-19 18:08:42 +08:00
Tianyi Cui bcc920369c docs: document package KV cache effects 2026-07-19 17:39:50 +08:00
Dudu-0223 a3f62478e1 Merge origin/master into codex/truncated-design 2026-07-17 18:21:54 +08:00
Dudu-0223 a4a9900be1 simplify retention omitted metadata 2026-07-09 13:53:51 +08:00
Dudu-0223 326b199f25 fix: address codex review round 2
- spill-policy reserves the spill notice's byte cost inside maxInlineBytes, so
  the replacement (preview + notice) never exceeds the documented model-facing
  cap. When the notice alone fills the budget the preview is empty; when even a
  notice-only replacement is not smaller than the original, the inline result is
  kept (spilling would only add bytes).
- retention TextRetainer trims an oversized single suffix chunk to the last
  suffixCap bytes on push, so tail/headTail retention stays bounded by suffixCap
  instead of retaining and re-copying the whole chunk in finish() — this is the
  spill preview path, which pushes the whole result as one chunk.
2026-07-09 09:51:35 +08:00
Dudu-0223 f4acb1cf05 fix: report omitted bytes against retained text, not budget (codex round 2)
finish() derived the exact omitted count from the pre-trim prefix/suffix
budget, but the UTF-8 boundary trims drop additional partial-codepoint bytes,
so an "Omitted N bytes" notice overstated what was kept (head maxBytes:2 over
'a€b' returned 'a' but claimed exact 3 when 4 bytes are absent). Derive the
count from the bytes actually returned (total − keptPrefix − keptSuffix) so
exact metadata matches the text and atLeast stays a valid lower bound.
2026-07-07 10:01:03 +08:00
Dudu-0223 cbfdf9d092 fix: preserve a codepoint spanning the head|tail split (codex round 1)
When headTail budgets cover the whole stream, omitted is 0 and the two
retained halves are contiguous — the split is artificial and a multibyte
codepoint can straddle it. finish() now decodes the contiguous buffer as one
in that case; the per-side UTF-8 boundary trims and separate decoding apply
only when a real middle gap exists. Without this, a headTail retainer could
drop a character while reporting truncated:false.
2026-07-07 09:46:15 +08:00
Dudu-0223 7a1c2779a7 feat(retention): add dsh-retention bounded-output library
Ship @deepseek-ai/dsh-retention under packages/util/: pure ItemRetainer /
TextRetainer plus neutral notice helpers, so tools that cap model-facing
output share one "what did we keep, what did we omit, may we stop reading"
mechanic while keeping grouping, exit codes, provider errors, and recovery
prose tool-owned. The two retainers are separate names because they differ in
resource model: item-head can stop the upstream on the first over-cap probe
(shouldStop), while text tail/head-tail must read to the end. The library
documents glob/grep/bash/web_fetch/web_search mappings but migrates no tool
yet — glob/grep don't exist, and migration is deliberately separate work.

Flips the RFC to implemented/ and rewrites its skeleton to shipped reality.
2026-07-07 09:33:59 +08:00