Commit Graph
6028 Commits
Author SHA1 Message Date
Tianyi Cui 02da51ff81 Merge branch 'codex/canonical-tool-output' into codex/code-mode-typed-results 2026-07-21 17:08:47 +08:00
Tianyi Cui 2276755c7f Merge branch 'codex/tool-json-schema-dsl' into codex/canonical-tool-output 2026-07-21 17:03:53 +08:00
Tianyi Cui 744eb99268 Merge branch 'master' into fix/compaction-kv-cache-reuse 2026-07-21 17:03:00 +08:00
Tianyi Cui a2c527308e Merge origin/master into worktree/docs-pages-deploy 2026-07-21 17:01:55 +08:00
Tianyi Cui 0c50ed9f8e Merge branch 'master' into codex/tool-json-schema-dsl 2026-07-21 16:59:17 +08:00
Tianyi Cui 7b63b85d24 Merge origin/master into fix/tui-color-scheme-v2 2026-07-21 16:58:12 +08:00
Tianyi Cui 35bb623cf4 Merge origin/master into worktree/routed-model-compaction-policy 2026-07-21 16:54:48 +08:00
Yichen Jiang 6fe8cfb5d1 Merge branch 'fix/acp-subagent-cwd' of github.com:deepseek-harness/deepseek-harness into fix/acp-subagent-cwd 2026-07-21 16:53:46 +08:00
Tianyi Cui 2434afd2ef Merge pull request #450 from deepseek-harness/codex/invariant-service-review-fixes
fix(invariants): harden runtime contracts and gates
2026-07-21 16:53:39 +08:00
Yichen Jiang 822651a2c7 test(subagent-acp): cover inherited cwd through a Loader composition
The inheritance branch ran only under hand-mounted plugin tests, and
the with-key e2e always configures an explicit cwd. Add a test-only
cordis.yml (scripted delegating model + the scripted mock ACP child)
booting the stdio app through the real Loader with cwd omitted: the
child proves it ran in — and was announced — the parent session's
workspace, keylessly.
2026-07-21 16:53:26 +08:00
Yichen Jiang f1e0410a5d Merge branch 'master' into worktree/semantic-session-checkpoints 2026-07-21 16:52:24 +08:00
Yichen Jiang 04117b58bd docs(subagent): align the parent contract with ACP cwd inheritance
SubagentStartRequest.parent, its core-data type-equiv mirror, and the
ACP backend Agent Note still said the out-of-process backend ignores
request.parent — no longer true: with cwd unconfigured it reads exactly
the session header's cwd, and a parent session without one now rejects
at start. State the one-field read at each owning contract, record the
workspace-resolution decision in the Agent Note, and align the README's
accessibility wording with the search-permission guard.
2026-07-21 16:51:49 +08:00
Yichen Jiang fcbbd22af7 fix(subagent-acp): require search permission on the child cwd
statSync().isDirectory() is true for a mode-600 directory, but a
subprocess cwd needs X_OK — spawn would fail EACCES, deferring a
self-contained configuration error past the load/start boundary the
resolution owns. Probe accessSync(X_OK) alongside the type check so
"not an accessible directory" is literally what the guard verifies.
2026-07-21 16:51:32 +08:00
Tianyi Cui 29ef3b5a7e Merge origin/master into fix/acp-subagent-cwd 2026-07-21 16:51:28 +08:00
Yichen Jiang 8394898ef5 test(tui): await async session suggestions 2026-07-21 16:50:45 +08:00
Tianyi Cui 2477ef415a Merge origin/master into lsp 2026-07-21 16:47:35 +08:00
Yichen Jiang 6cd16fd424 Merge origin/master into worktree/session-reference 2026-07-21 16:47:04 +08:00
Yichen Jiang 32d786c439 feat(session): add cross-session references 2026-07-21 16:46:48 +08:00
Tianyi Cui b9913db7de Merge pull request #459 from deepseek-harness/worktree/pr329-review-fixes
fix(lsp): align operation names and review contracts
2026-07-21 16:34:50 +08:00
Tianyi Cui 5261ebab38 fix(compact-basic): fence surface changes only 2026-07-21 16:34:28 +08:00
Tianyi Cui 13982e98e1 Merge pull request #458 from deepseek-harness/fix/pre-push-dsh-home-env-leak
test(workspace-context): stub DSH_HOME in default-home label test
2026-07-21 16:16:45 +08:00
NI0317 85ac747208 fix: wait for PTY startup readiness 2026-07-21 16:12:42 +08:00
Tianyi Cui 6475e51825 fix(goal-session): discriminate plugin turn triggers 2026-07-21 16:08:53 +08:00
NI0317 58cde5103a feat: add persistent PTY sessions 2026-07-21 16:01:00 +08:00
Turtle e97290ba9e docs(telemetry): state anonymous id as per-harness-home, not per-machine
The consolidated resolver scopes the anonymous id to $DSH_HOME rather
than the machine. Update the module contract, README, and Agent Note to
say per-harness-home explicitly instead of over-claiming a machine-global
identity, and record why DSH_HOME scoping is the intended single-root
meaning rather than a regression.
2026-07-21 15:57:17 +08:00
Tianyi Cui 5d411f9c4e fix(session-title): preserve agent turn outcomes 2026-07-21 15:49:37 +08:00
Yichen Jiang ddc1c66c03 fix(python): mount session checkpoint policy 2026-07-21 15:23:45 +08:00
Tianyi Cui c622a2881d fix(session-title): deduplicate fallback creation 2026-07-21 15:23:27 +08:00
Tianyi Cui 9b048354b8 fix(session-title): enforce framed input and deadline 2026-07-21 15:00:15 +08:00
Yichen Jiang 6d12e3ab41 fix(session): add semantic crash checkpoints 2026-07-21 14:50:06 +08:00
Yichen Jiang 49a9bca9f6 fix: validate compaction config pairs 2026-07-21 14:49:57 +08:00
Turtle fac3fc090f fix(paths): treat empty DSH_HOME as unset; isolate telemetry env test
Review fixes for #462:
- resolveDshHome now treats an empty or whitespace-only $DSH_HOME as
  unset, so a blank override never resolves the home to cwd via
  resolve(''). Restores the guard telemetry's old resolver carried.
- The default-env telemetry test asserts only that globalConfigDir()
  returns an absolute path, so a machine DSH_HOME without a .dsh suffix
  cannot break it.
2026-07-21 14:43:38 +08:00
Tianyi Cui d9557fa5e0 fix(session-title): serialize provider request writes 2026-07-21 14:39:09 +08:00
Tianyi Cui b3ba4345f5 fix(session-title): respect request and config boundaries 2026-07-21 14:09:17 +08:00
Wenlu Wang 317f740dc2 Merge pull request #447 from deepseek-harness/codex/pr239-plan-mode-review-fixes
fix(mode): stabilize plan-mode model experience
2026-07-21 14:01:41 +08:00
Turtle a0268d2509 fix(compact-basic): replay conversation prefix so summarization reuses KV cache
Automatic compaction fires mid-conversation, right after the loop warmed
the provider's KV cache with the last routed request. The default
summarizer then issued a separate request whose prefix shared nothing
with that warm request — a bespoke summarizer system prompt followed by
the older history flattened to one rendered transcript string — so a
differing first token invalidated the entire cached prefix and every
compaction re-processed the whole replayed history twice.

Move the compaction directive from the FRONT (a fresh system prompt) to
the END (a trailing user message), and replay the last routed request's
own system prompt, tools, message prefix, and shadowed-region messages
verbatim via session.requestHeader() + deriveEventMessage. The auxiliary
call is now a genuine prefix-extension of the warm request, so the
provider reuses the cached tokens up to the trailing instruction.

SummarizationInput carries the replayed prefix instead of a flat string;
the now-unused renderTranscript/renderContentBlocks path is removed with
its spec. Cache reuse is best-effort (head compaction guarantees a hit;
a mid-range compaction or a differently-routed summarizer forgoes it),
correctness is not.
2026-07-21 13:55:22 +08:00
Tianyi Cui bc60b9ffe8 fix(lsp): cancel blocked document opens 2026-07-21 13:52:15 +08:00
Turtle 92e0d0e04f refactor(paths): collapse harness home resolution into one resolver
Delete @deepseek-ai/dsh-home and make dsh-paths the sole owner of the
single-root harness home ($DSH_HOME || ~/.dsh). Migrate tool-bash,
skill-local, and agent-spine-demo off dsh-home, and fold telemetry's
divergent globalConfigDir onto the shared resolver, dropping its second
XDG/APPDATA policy and the deepseek-harness namespace so the anonymous
id lives under the harness home. Add dshHomeDisplay() for symbolic
user-facing paths, replacing workspace-context's bespoke check.
2026-07-21 13:52:00 +08:00
Tianyi Cui 63cb16a540 fix(session-title): honor request lifecycle 2026-07-21 13:37:28 +08:00
Tianyi Cui 2fd995bf3c fix(lsp): align operations and harden lifecycle 2026-07-21 13:29:40 +08:00
Huanqi Cao 5d6db0cd56 fix: stub DSH_HOME in 'labels the default DSH home' test
The test 'labels the default DSH home as ~/.dsh when HOME points at the
configured default' mocks node:os to return a temporary home directory
and expects discoverBaselineInstructionFiles to fall back to the default
~/.dsh path. However, when DSH_HOME is externally set (e.g. in the DSH
agent environment), resolveDshHome(undefined) reads that external value
instead of falling back to defaultDshHome(), causing the test to look for
AGENTS.md in the wrong directory and return an empty result.

Stub DSH_HOME to undefined so the test controls the full resolution path.
2026-07-21 13:08:34 +08:00
Tianyi Cui d6bebc18f8 Merge remote-tracking branch 'origin/codex/enforce-tool-cancellation' into worktree/explicit-turn-signal
# Conflicts:
#	docs/architecture.md
#	docs/cordis-catalog/events.md
#	docs/core-data-structures/core.md
#	docs/event-producer-consumer.md
#	packages/core/agent-loop/README.md
#	packages/core/agent-loop/src/agent.ts
#	packages/core/agent/README.md
#	packages/core/agent/src/types.ts
#	packages/ui/acp/src/index.ts
#	packages/ui/tui/src/index.ts
#	packages/ui/tui/tests/harness.ts
2026-07-21 12:48:46 +08:00
Tianyi Cui 43d50a38b1 Merge remote-tracking branch 'origin/master' into codex/enforce-tool-cancellation
# Conflicts:
#	docs/architecture.md
#	packages/core/agent-loop/README.md
2026-07-21 12:30:08 +08:00
Yichen Jiang 7e43f8661c Merge branch 'master' into fix/acp-subagent-cwd 2026-07-21 12:29:50 +08:00
Tianyi Cui c6e1d35a99 fix(core): close turn cancellation contract gaps 2026-07-21 12:14:53 +08:00
Tianyi Cui a9d518a38e fix(session-title): harden async provider lifecycle 2026-07-21 12:08:00 +08:00
Tianyi Cui 06e4932984 Merge latest #239 into #447
# Conflicts:
#	.agents/notes/implemented/feature/2026-07-07-plan-mode.md
#	docs/config-catalog.md
#	docs/cordis-catalog/services.md
#	packages/mode/README.md
#	packages/mode/mode/README.md
#	packages/mode/mode/src/index.ts
#	packages/mode/mode/tests/mode.spec.ts
2026-07-21 11:43:54 +08:00
Chinesezjc 6c34d69c51 Merge remote-tracking branch 'origin/master' into chore/doc-sync-single-pass
# Conflicts:
#	docs/development.i18n.yaml
#	package.json
2026-07-21 11:29:49 +08:00
kingwl a00d1e0496 refactor(mode): move exit-tool visibility onto the registry restriction layer; register /mode; drop unconsumed surfaces
The master merge brought the tool registry's per-scope restriction layer
(tools.restrict), which makes dsh-mode's prepend assemble filter a duplicate
enforcement shape: it re-implemented the registry's SDK-section rendering
(renderToolsSdk + the RUN_CODE_NAME exclusion) and hid the exit tool from
prompts only — dispatch stayed open and the execute-time re-check was the
real gate. The service now reconciles a per-agent deny restriction on
agent.ctx at agent/created and at every boundary flush, so wire schemas, the
Code Mode tools:sdk section, AND dispatch resolve exit_plan_mode through the
one registry view (a default-mode call answers UNKNOWN_TOOL, byte-identical
to a no-dsh-mode deployment). The execute-time folded-mode re-check stays as
defense in depth for a direct foreign mode/set append no boundary has
reconciled yet.

Two zero-consumer surfaces removed per the pre-release stance:
- AgentOptions.mode creation seeding (declaration merge + agent/created
  listener); a caller selects through set() before the first turn, and the
  deferred subagent inheritance returns together with its consumer.
- The dropped-definition boundary notice (droppedNoticed + narration):
  custom mode definitions have no production consumer, so nothing can be
  dropped; fold-to-default degradation is unchanged.

The stdio removal had left plan mode ACP-only while docs still claimed a
/mode command. dsh-mode now registers /mode on the plugin-owned command
registry through an optional ctx.inject(['commands']) child (type-only peer
edge on dsh-commands), so the TUI and the ACP slash-command surface both
gain it; examples/tui-agent composes dsh-mode. ACP/TUI expected outputs
refreshed keyless for the available_commands_update delta.

Docs updated in place (mode READMEs, the plan-mode Agent Note's realization
sections); catalogs and graphs regenerated.
2026-07-21 11:11:35 +08:00
Chinesezjc 6fc7dd4c02 chore(gates): route doc-sync through the run-gates scheduler
pnpm run doc-sync was a serial &&-chain of 24 pnpm run subcommands; the
24 script bodies finish in ~34s while the chain takes ~3min, all wrapper
overhead. Add a doc-sync mode to scripts/run-gates.ts expanding to
docSyncLeafGates() (concurrency capped at 4 like pre-push: several doc
gates each build a full ts.Program) and point the package script at it,
making the leaf list the single source of truth for the member set.

Consolidating the two lists surfaced drift: verify-cordis-api joined the
package.json chain with the runtime API catalog but was never added to
docSyncLeafGates, so CI and pre-push never gated that catalog's
freshness. Add the missing leaf.
2026-07-21 11:07:50 +08:00