Commit Graph

989 Commits

Author SHA1 Message Date
Hypatia May
4f55c15cf8 Merge remote-tracking branch 'origin/master' into session-fork 2026-07-07 09:06:00 +08:00
Hypatia May
c7fba41ba9 Inline session fork parameters 2026-07-07 09:04:49 +08:00
Tianyi Cui
81cbc3daa4 Merge pull request #191 from deepseek-harness/catalog-flatten
docs: flatten single-file catalog dirs to docs/tool-catalog.md and docs/persistence-catalog.md
2026-07-07 00:30:36 +08:00
Tianyi Cui
0c568201c9 Merge remote-tracking branch 'origin/master' into catalog-flatten 2026-07-07 00:29:12 +08:00
Tianyi Cui
215ff9a4e6 Merge pull request #188 from deepseek-harness/config-catalog
feat: generated plugin config catalog (docs/config-catalog.md)
2026-07-07 00:27:59 +08:00
Tianyi Cui
d45e470192 fix: retarget the i18n pairing exclusions to the flattened catalog paths
The manifest excluded docs/tool-catalog/ and docs/persistence-catalog/ as
directories; the trailing slash is a deliberate path boundary, so the
flattened single files no longer matched and lost their generated-doc
classification — the gate would have accepted a hand-added .zh.md twin it
must reject. The entries now name the files.
2026-07-07 00:08:55 +08:00
Tianyi Cui
4e2da27201 Merge branch 'config-catalog' into catalog-flatten 2026-07-07 00:06:19 +08:00
Tianyi Cui
f0ef280848 fix: identity-check name resolutions across the pasted closure; exclude the catalog from i18n pairing
Review findings on the config catalog:

The transitive-paste closure deduplicated by bare name, so two DIFFERENT
package-local declarations sharing a name would silently collapse to the
first one pasted — and a reference satisfied by the wrong declaration is a
wrong catalog, not just a thin one. Every resolution is now identity-checked
by source pointer: reaching the same declaration again is benign, while a
name that resolves to two distinct declarations, to conflicting imports, or
to a declaration in one file and an import in another is a violation — a
verbatim fence has a single flat namespace, so the fix is a rename at the
source. A spec case pins the two-declarations collision.

docs/config-catalog.md also joins the translation-pairing exclusions: it is
a generated English-only doc, and without the entry the gate would accept a
hand-added .zh.md twin it must reject.
2026-07-07 00:00:34 +08:00
Tianyi Cui
82dfc62620 Merge branch 'config-catalog' into catalog-flatten 2026-07-06 23:38:40 +08:00
Tianyi Cui
fe815c9586 Merge remote-tracking branch 'origin/master' into config-catalog
# Conflicts:
#	docs/rfc/INDEX.md
2026-07-06 23:31:58 +08:00
Tianyi Cui
6d9a4f1258 Merge branch 'config-catalog' into catalog-flatten
# Conflicts:
#	docs/config-catalog.md
#	scripts/gen-config-catalog.ts
2026-07-06 23:16:29 +08:00
Tianyi Cui
fb3f3a69ec fix: extend the schema cross-check to nested key paths; scope the catalog framing
Review findings on the config catalog:

The schema-subset check compared only top-level z.object keys against
top-level type members, so a nested loader-accepted key (agents[].id,
agentOptions.model, capabilities.*) missing from the declared type would
pass the gate unseen. The walk now collects nested object/array
compositions as key paths and resolves each against the declared config
type — through interfaces (heritage included), aliases, literals,
intersections, unions, arrays, indexed access, Partial-style wrappers, and
type references across package-local and workspace imports (re-export
chains included). The check stays presence-only and one-directional, and
only a definite miss is a violation: a path crossing a type the walk
cannot enumerate (an external package's) is skipped, never mis-reported.
The recursion guard applies at named declarations only — a structural
first child shares its span start with its parent, so a span-keyed guard
on every node mistakes ordinary descent for a cycle and silently turns
definite misses into unknowns.

The page and RFC framing also overstated the catalog as the exact
cordis.yml-settable surface: the paste is the plugin's full declared
config type, and a field the runtime schema deliberately excludes (the ACP
bridge's test-injected stream) is a runtime-only seam its own JSDoc marks.
Both now say so.

Five spec cases pin the new behavior: nested hidden key, workspace
intersection via star re-export, Partial wrapper, indexed-access
composition, and the external-type unknown path.
2026-07-06 23:06:03 +08:00
Tianyi Cui
116e2c45a8 docs: flatten single-file catalog dirs to docs/tool-catalog.md and docs/persistence-catalog.md
A directory holding exactly one file adds a path level for nothing;
cordis-catalog/ keeps its folder because it holds two sibling pages
(events.md + services.md), matching config-catalog.md which was born flat.

docs/tool-catalog/tools.md -> docs/tool-catalog.md and
docs/persistence-catalog/log-events.md -> docs/persistence-catalog.md; their
generators' OUT paths and in-page relative links drop one directory level,
and every citation repo-wide (docs, RFCs, package READMEs, generator
headers) is updated. graph-atlas.md, config-catalog.md, and the doc graphs
regenerate with the new paths.
2026-07-06 22:26:06 +08:00
Tianyi Cui
0606cd559c Merge pull request #175 from deepseek-harness/codex/parallelize-ci
ci: parallelize GitHub workflows
2026-07-06 22:15:13 +08:00
Tianyi Cui
8458530298 Merge branch 'master' into codex/parallelize-ci 2026-07-06 22:12:44 +08:00
Wenlu Wang
654d192c47 Merge pull request #182 from deepseek-harness/codex/gate-cadence-docs
docs: clarify local gate cadence
2026-07-06 22:10:19 +08:00
Tianyi Cui
b0c7eadd23 feat: generated plugin config catalog (docs/config-catalog.md)
scripts/gen-config-catalog.ts walks every packages/<group>/<pkg> entry with
the TypeScript compiler API and emits docs/config-catalog.md: per loadable
plugin, the verbatim config declaration (JSDoc included) its apply/constructor
receives in a ts config-catalog fence, the inject requirements, resolved links
for every referenced type (package-local types pasted transitively, other
plugins' config types as intra-page anchors, LINK_MAP names to
core-data-structures, workspace types to source), and terse classification
lists for config-free plugins, abstract seams, and libraries — classification
is total, so a new package cannot go undocumented.

The walk enforces per-field JSDoc prose on every pasted declaration and
statically cross-checks the schemastery schema (z.object keys, z.intersect
composition across packages): every schema-validated key must be a declared
member of the config type. One violation existed repo-wide — the agents[].id
field in dsh-agent-loop — fixed by adding its JSDoc (which shifts the
cordis-catalog services page's source pointers; regenerated).

verify-config-catalog (--check) joins doc-sync; doc-typecheck learns the
ts config-catalog fence; gen-cordis-catalog exports its JSDoc/pointer helpers
and LINK_MAP for reuse. Negative-path spec in
packages/core/agent-core/tests/gen-config-catalog.spec.ts mirrors the
gen-cordis-catalog spec. Decision record:
docs/rfc/implemented/process/2026-07-06-generated-config-catalog.md (includes
the deliberate acceptance of README ## Config overlap).
2026-07-06 21:57:17 +08:00
Tianyi Cui
9df79534a6 Merge branch 'master' into codex/gate-cadence-docs 2026-07-06 21:55:48 +08:00
Tianyi Cui
c57c7527a5 Merge branch 'master' into codex/parallelize-ci 2026-07-06 21:54:07 +08:00
Tianyi Cui
1c999804d8 Merge pull request #184 from deepseek-harness/codex/clarify-doc-budget-guidance
Clarify doc budget guidance
2026-07-06 20:46:31 +08:00
kingwl
e48a344f6a docs: keep gate cadence guidance agent-only 2026-07-06 17:08:50 +08:00
Hypatia May
41e1955ace Clarify condensing in doc budget guidance 2026-07-06 16:31:18 +08:00
Hypatia May
9c5da8d81f test: cover corrupt fork boundary guard 2026-07-06 16:16:47 +08:00
Hypatia May
72166dad03 Clarify doc budget guidance 2026-07-06 16:11:56 +08:00
Hypatia May
50ffa0c9a9 test: use portable command in fs policy snapshot 2026-07-06 16:00:52 +08:00
Hypatia May
6a7bb4045b test: normalize platform sed snapshot output 2026-07-06 14:35:22 +08:00
Hypatia May
aed594813b Merge remote-tracking branch 'origin/master' into session-fork
# Conflicts:
#	docs/architecture.md
#	docs/cordis-catalog/services.md
2026-07-06 14:30:36 +08:00
Hypatia May
71f71816b9 fix: keep fork boundary check simple 2026-07-06 14:17:31 +08:00
Hypatia May
37aac7f313 fix: collapse session fork to one api 2026-07-06 13:57:59 +08:00
kingwl
e1b85da182 docs: trim daily command wording 2026-07-06 13:45:17 +08:00
kingwl
7af8f98080 docs: clarify local gate cadence 2026-07-06 13:45:17 +08:00
Tianyi Cui
cbc9c26366 Merge pull request #179 from deepseek-harness/worktree-reconstructable-requests
Make every LLM request a pure function of the session log
2026-07-06 13:37:20 +08:00
Hypatia May
37f3aedc0b docs: keep architecture summary within budget 2026-07-06 12:37:54 +08:00
Hypatia May
0bf8749128 fix: fold session fork into session store 2026-07-06 12:35:34 +08:00
Hypatia May
cf036299f3 Merge remote-tracking branch 'origin/master' into session-fork
# Conflicts:
#	docs/architecture.md
#	docs/cordis-catalog/events-and-services.md
#	docs/core-data-structures/core.md
#	docs/module-graph.md
#	docs/rfc/README.md
#	packages/README.md
#	tsconfig.base.json
2026-07-06 12:20:00 +08:00
Tianyi Cui
39f2898502 fix review findings: last stale compact README line; stale-mark the Code Mode proposal
The compact-basic README's opening line still described summarization
as routed through the agent request pipeline — corrected to the direct
one-shot llm/stream contract the rest of the package documents. The
proposed Code Mode RFC gets a premise-stale note: it names the old
full-request agent/request seam, which now shapes call config only —
re-map onto the log channels and system-prompt/assemble before
implementing from it.
2026-07-06 04:22:37 +08:00
Tianyi Cui
8bd80e5e9b fix review findings: one frozen seed through the waterfall; stale agent/request docs
Codex diff review, round 1, two (A) findings:

- The agent/request fallback resolved the RAW seed object — on later
  steps the session's cached header fold — so a delegating listener
  (await next(), mutate, return) could rewrite the fold in place and
  the change would compare as already-baseline: no delta logged, the
  persisted log unable to reconstruct the request (the dev invariant
  would fire on the divergence, but the log would still lie). One
  structuredClone'd, deep-frozen seed now serves both the listener
  chain and the fallback — in-place shaping after delegation throws —
  and Session.requestHeader() freezes its fold on update, so the leak
  class is unrepresentable from either side. Pinned by a loop-level
  delegating-mutator test.
- Doc sweep for the old contract: agent README's event row (mutate
  GenerateOptions / tool filtering → frozen config seed, replacement
  out, logged header), compact-basic's module JSDoc (summarize routed
  through agent/request → direct one-shot at llm/stream), and
  architecture.md's event-domain line (request mutation → call-config
  shaping).
2026-07-06 04:17:29 +08:00
Tianyi Cui
23ed47ba75 chore: knip learns agent-loop's e2e entry (ported from #162 with request-cache.e2e.ts) 2026-07-06 04:01:56 +08:00
Tianyi Cui
c58e5adee7 test: close the coverage gaps the gate found
surface.replaceGeneration gets its direct-access test (the getter folds
a pending delta itself, not only via a nodes read); the reconstruction
theorem drives maxTokens and stop through the config waterfall; the
compact-basic envelope drops a dead conditional — config.maxTokens is
required and validated positive, so this backend's envelope always
carries the cap (the return type's optionality exists for overriding
subclasses).
2026-07-06 03:55:30 +08:00
Tianyi Cui
c0808d5126 docs: the governing principle — every LLM request is reconstructable from the session log
The reconstructability RFC is the principle's home: model-visible ⟺
logged in both forms, the mechanism (boundary derivation + header
fold), the enforcement (write-time round-trip guard, the dev
invariant), the corollaries ranked (prefix-cache stability first), the
MiniCode lineage with the provenance arrow inverted, and the
alternatives it beat — including the stateful transmission client
whose three-design archaeology lives in PR #162.

Placements per the one-home-per-fact taxonomy: a standing-order line in
root AGENTS.md (with displacement trims to stay inside the 1,575-word
ceiling), the principle statement in architecture.md § Session Log and
its Turn Flow lines (condensed to the ratcheted 1,630 ceiling), the
request-envelope section in core-data-structures/core.md with the
LlmCallConfig paste, both review-requested FIXMEs
(FIXME(call-config-shape) beside the type, FIXME(catalog-verbs) at the
catalog's drift-gate note), cookbook rows redirected off agent/request
(tool filtering → system-prompt/assemble, plan-mode prompt → sections/
inject()), and the llm/stream JSDoc stating the frozen-request
contract. RFC index and all generated catalogs regenerated.
2026-07-06 03:49:35 +08:00
Tianyi Cui
7376346246 test: re-record snapshot goldens — every transcript gains its request/header event
One re-record after the header events landed: recorded scenarios
re-harvested against the live API; the three fs-writing fixtures are
relativized (this recording's model happened to emit absolute
file_path arguments, which only round-trip through replay when the
path is cwd-independent) and, with the two never-re-recorded authored
scenarios (error-finish, cancel), rewritten in the normalized
authored-fixture form the harness documents — each now carrying the
request/header snapshot the loop logs before its first dispatch.
Keyless replay verified green across all 35 scenarios.
2026-07-06 03:39:48 +08:00
Tianyi Cui
7539231eca compact: the summary's provenance records its call envelope
compact/summary gains { model, maxTokens? } — the envelope the
summarize call actually used, reported by the backend that made the
call: summarize() now returns { summary, model, maxTokens? } instead of
bare blocks, so an overriding backend (template or remote summarizer)
reports its own envelope honestly and compactRegion logs it. 'Which
model wrote this summary' becomes answerable from the log alone, and
the one-shot summarize request — outside the loop's header-event fold
by design — is reconstructable from log + code (the reconstructability
RFC's scope statement).
2026-07-06 03:21:29 +08:00
Tianyi Cui
616c2ffba3 ci: keep measured e2e worker cap 2026-07-06 03:20:08 +08:00
Tianyi Cui
cc670c0762 ci: use full e2e worker fanout 2026-07-06 03:16:34 +08:00
Tianyi Cui
65d644b161 invariants: a loop-built request must be exactly what the log reconstructs
The dev-mode cross-check on llm/stream: a frozen request with a live
sessionId (the loop-built marker; hand-built one-shots stay unfrozen
and skipped) must carry messages deep-equal to the derivation over the
log prefix strictly before the in-flight step's step/start, and header
fields equal to the fold of the log's request/header* events. The
messages side rebuilds through a FRESH Session over the boundary
prefix — same projection code, zero shared state — so the live cache
cannot vouch for itself, and the seq-bounded rebuild is boundary-
correct: content appended after step/start (an agent/request-window
inject) legitimately belongs to the next request and cannot false-fire
the check. prepend:true only defends against the replay adapter's
short-circuit (append-registered); correctness never rests on listener
ordering. There is no divergence-allowance caveat: nothing can shape
request content outside the log.
2026-07-06 03:14:07 +08:00
Tianyi Cui
3464f88297 test: shorten compaction e2e smoke 2026-07-06 03:12:37 +08:00
Tianyi Cui
1b9408268b ci: cache eslint lane 2026-07-06 03:07:35 +08:00
Tianyi Cui
2093a8898b loop: every request is built from the log — boundary snapshot, header events, config-only waterfall
The loop is now transmission-stateless; a request is a pure function of
(session log, this step's rendered assembly, current AgentOptions):

- The reconstruction boundary is step/start: the messages snapshot is
  taken in the same synchronous frame immediately before the step/start
  append, so the request's messages are exactly the derivation over
  events[0..stepStartSeq) — an inject() from an agent/request listener
  (or any concurrent task) lands after the boundary and joins the NEXT
  request. This changes behavior for a synchronous step/start
  session/event listener that appends content (master derived after the
  append, so such a listener could reach the current request):
  agent/pre-step is the sanctioned seam for current-request content.
- agent/request is re-typed to config-only: (agent, turn, step,
  config: LlmCallConfig, next) → LlmCallConfig. The frozen seed comes
  from AgentOptions on a loop instance's first request (explicit options
  beat the logged baseline — fork overrides and resume reconfiguration
  stay correct) and from the log's folded header afterwards; listeners
  return a replacement to switch. Content shaping through the request is
  no longer expressible — model-visible content flows through the log
  channels.
- recordRequestHeader appends whatever header event the request owes the
  log before dispatch: an 'initial'/'resume' snapshot anchoring each
  loop instance, a round-trip-verified delta on change, a 'fallback'
  snapshot when the encoding cannot express it. Session.requestHeader()
  is the log's incrementally-folded baseline.
- Requests are deep-frozen before dispatch (deepFreeze exempts the
  AbortSignal — freezing one breaks AbortController.abort() outright);
  frozen + sessionId is the loop-built marker the dev invariant keys on.

Ported from #162 and re-anchored on the log: the append-extension /
frozen-end-to-end / compaction-resend / prompt-change property tests,
plus new specs for the boundary semantics, resume anchoring, and the
end-to-end theorem (every recorded request rebuilds byte-equal from the
log alone). Live cache-hit e2e (request-cache.e2e.ts) verified against
the real DeepSeek API. Snapshot goldens intentionally stale until the
single re-record after the compact/summary envelope lands.
2026-07-06 03:07:34 +08:00
Tianyi Cui
d71c96a44c ci: keep lint lane serial 2026-07-06 02:59:50 +08:00
Tianyi Cui
09326fd84a ci: split lint lane internally 2026-07-06 02:56:23 +08:00