Hunter Bown
2b915b687f
Merge remote-tracking branch 'origin/pr/1197' into work/v0.8.34
...
# Conflicts:
# crates/tui/src/project_context.rs
2026-05-12 23:17:53 -05:00
Hunter Bown
b8036b4b4a
Merge remote-tracking branch 'origin/pr/1346' into work/v0.8.34
2026-05-12 23:17:36 -05:00
Hunter Bown
b21c733934
Merge remote-tracking branch 'origin/pr/1270' into work/v0.8.34
...
# Conflicts:
# crates/tui/src/working_set.rs
2026-05-12 23:17:34 -05:00
Hunter Bown
6732e009e4
Merge remote-tracking branch 'origin/pr/1203' into work/v0.8.34
2026-05-12 23:16:52 -05:00
Hunter Bown
52b70afa3b
Merge remote-tracking branch 'origin/pr/1211' into work/v0.8.34
...
# Conflicts:
# crates/tui/src/tui/app.rs
# crates/tui/src/tui/ui.rs
# crates/tui/src/tui/ui/tests.rs
2026-05-12 23:16:50 -05:00
Hunter Bown
87de4e8744
Merge remote-tracking branch 'origin/pr/1230' into work/v0.8.34
2026-05-12 23:16:13 -05:00
Hunter Bown
7c720c9841
Merge remote-tracking branch 'origin/pr/1233' into work/v0.8.34
...
# Conflicts:
# crates/config/src/lib.rs
2026-05-12 23:16:10 -05:00
Hunter Bown
36ed661dc8
Merge remote-tracking branch 'origin/pr/1235' into work/v0.8.34
2026-05-12 23:15:50 -05:00
Hunter Bown
cf62ab65e3
Merge remote-tracking branch 'origin/pr/1256' into work/v0.8.34
...
# Conflicts:
# crates/tui/src/mcp.rs
2026-05-12 23:15:45 -05:00
Hunter Bown
a415359577
style: format v0.8.34 PR integrations
2026-05-12 23:14:49 -05:00
Hunter Bown
88b7882d48
Merge remote-tracking branch 'origin/pr/1432' into work/v0.8.34
2026-05-12 23:14:24 -05:00
Hunter Bown
ad297f4de1
Merge remote-tracking branch 'origin/pr/1500' into work/v0.8.34
2026-05-12 23:14:22 -05:00
Hunter Bown
a7df269832
Merge remote-tracking branch 'origin/pr/1505' into work/v0.8.34
2026-05-12 23:14:19 -05:00
Hunter Bown
b6308620b4
Merge remote-tracking branch 'origin/pr/1508' into work/v0.8.34
...
# Conflicts:
# crates/tui/src/tui/ui.rs
2026-05-12 23:14:17 -05:00
Hunter Bown
354dd92889
Merge remote-tracking branch 'origin/pr/1513' into work/v0.8.34
2026-05-12 23:13:57 -05:00
Hunter Bown
e7fba0af3e
Merge remote-tracking branch 'origin/pr/1504' into work/v0.8.34
2026-05-12 23:13:31 -05:00
Hunter Bown
0076e59a82
Merge remote-tracking branch 'origin/pr/1514' into work/v0.8.34
2026-05-12 23:13:29 -05:00
Hunter Bown
156703ae59
Merge remote-tracking branch 'origin/pr/1518' into work/v0.8.34
2026-05-12 23:13:27 -05:00
Hunter Bown
f4f1916221
Merge remote-tracking branch 'origin/pr/1548' into work/v0.8.34
2026-05-12 23:13:25 -05:00
Hunter Bown
81e4b93cc9
Merge pull request #1553 from Hmbown/work/v0.8.33
...
chore(release): prepare v0.8.33
2026-05-12 22:30:53 -05:00
Hunter Bown
a507885fb8
test(rlm): tolerate windows python line endings
2026-05-12 22:18:59 -05:00
Hunter Bown
3c3c4bf824
fix(ci): satisfy release lint and isolate config tests
2026-05-12 22:10:03 -05:00
Hunter Bown
485ba7bbd4
chore(release): finish v0.8.33 polish
2026-05-12 22:03:47 -05:00
Hunter Bown
c7e5cb4160
test: fix three tests that read host settings.toml default_mode
...
The user's ~/Library/Application Support/deepseek/settings.toml had
default_mode = "yolo", which caused test_mode_yolo_sets_all_flags,
test_trust_on_enables_flag, and
footer_status_line_spans_show_mode_and_model_idle_and_active to fail
because they implicitly depended on the host's global mode setting.
Pin each test to Agent mode explicitly so they pass regardless of the
developer's personal settings.
2026-05-12 21:59:11 -05:00
Hunter Bown
99c6b22e83
chore(release): v0.8.33 — sub-agent and RLM renovation with persistent sessions
...
- Persistent RLM sessions (rlm_open/rlm_eval/rlm_close) with bounded REPL helpers
- Fork-aware sub-agent sessions (agent_open/agent_eval/agent_close) with handle_read
- Shared handle_read storage with slice/range/count/JSONPath projections
- Slash-command routing: /rlm, /agent, /relay (/接力) for handoff prompts
- Sidebar renamed to "Work" tab, consistent across Plan/Agent/YOLO modes
- Tool papercuts: file_search excludes, grep_files strings, fetch_url JSON,
edit_file fuzz, exec_shell merged stdout/stderr, revert_turn no-op reject
- CLI reasoning-effort honoured on non-auto exec routes (#1511 @h3c-hexin)
- Edit-file replacement boundaries clarified (#1516 )
- Pandoc output validated before probing (#1523 )
- Running turns steerable/repaintable (#1533 , #1537 )
- Tasks/Activity Detail calmer under load
- npm retry timeout hint (#1538 @reidliu41)
- Issue templates improved (#1525 @reidliu41)
- Shell: kill process group to prevent UI freeze (#828 @CrepuscularIRIS)
- TUI: ignore leaked SGR mouse reports in composer (#1421 @reidliu41)
- Footer: keep chips within available width (#1417 @Wenjunyun123)
- Session picker: scope Ctrl+R to current workspace (#1395 @LinQ)
- Removed stale competitive-analysis doc
- Prompts/docs teach only new tool names
2026-05-12 19:54:08 -05:00
zhuang biaowei
2346252a63
Fix changelog command version hints
2026-05-13 08:17:19 +08:00
Reid
503551ddec
docs: improve issue templates ( #1525 )
...
Add clearer prompts to bug reports and feature requests so users provide
reproduction details, impact, environment context, use cases, and supporting
materials up front.
2026-05-12 15:37:11 -05:00
hexin
ec527b6a2b
fix(cli): honor config.toml reasoning_effort on non-auto exec routes ( #1511 )
...
`resolve_cli_auto_route` was hard-coding `reasoning_effort: None` when
`--model` is not `auto`, which silently dropped the value the user had
set in `~/.deepseek/config.toml` on every non-auto-route exec/one-shot
call.
For vllm + Qwen3 users with `reasoning_effort = "off"`, thinking was
therefore never disabled. The model emitted a long reasoning trace for
every prompt and SSE idle timeouts (`did not receive response headers
after 45s`) fired on any non-trivial prompt. After this fix, the same
prompts return in ~1.5s.
Route the configured value through `ReasoningEffort::from_setting`, the
same parser the TUI uses elsewhere for this field. Auto-route behaviour
(`--model auto`) is unchanged.
Verified by capturing the outgoing request body with `nc` before and
after; chat_template_kwargs.enable_thinking=false now appears in the
body on vllm exec runs.
Co-authored-by: hexin <he.xin@h3c.com >
2026-05-12 15:37:04 -05:00
Hunter Bown
9fb3d5d636
fix(web): keep contributor stats current
2026-05-12 15:32:01 -05:00
Hunter Bown
ad70655bbb
docs(release): note v0.8.32 selection issue
2026-05-12 15:08:21 -05:00
Hunter Bown
3a1b107af9
chore(release): pin security contact and cnb tag sync
2026-05-12 14:48:10 -05:00
Hunter Bown
b7f14b2116
fix(release): package changelog with tui crate
2026-05-12 14:34:17 -05:00
Hunter Bown
2326220b7e
fix(vision): reject rooted image paths on windows
2026-05-12 14:13:19 -05:00
Hunter Bown
190eb6b162
docs(web): refresh v0.8.32 site state
2026-05-12 14:04:17 -05:00
Hunter Bown
b25450728e
fix(docker): include changelog in image build context
2026-05-12 13:01:16 -05:00
Hunter Bown
8a76916140
docs(changelog): finalize v0.8.32 notes
2026-05-12 12:10:44 -05:00
Hunter Bown
2662f36001
docs: fix working set rustdoc links
2026-05-12 12:10:39 -05:00
Hunter Bown
9ca759bd11
fix(npm): show timeout hint on first retry ( #1538 )
...
Co-Authored-By: jieshu666 <jieshu666@users.noreply.github.com >
2026-05-12 12:10:29 -05:00
Hunter Bown
c535f85a4d
fix(tools): clarify edit_file replacement boundaries ( #1516 )
2026-05-12 12:10:22 -05:00
Hunter Bown
136be0d255
test(tui): pin model state in footer and picker tests
2026-05-12 12:10:17 -05:00
Hunter Bown
92fd81db4e
fix(pandoc): validate binary output before probing pandoc ( #1523 )
...
Co-Authored-By: muyuliyan <muyuliyan@users.noreply.github.com >
2026-05-12 12:10:11 -05:00
Hunter Bown
ceac2aa776
fix(tui): steer and repaint running turns ( #1533 , #1537 )
...
Co-Authored-By: Oliver-ZPLiu <Oliver-ZPLiu@users.noreply.github.com >
Co-Authored-By: czf0718 <czf0718@users.noreply.github.com >
2026-05-12 12:10:04 -05:00
Hunter Bown
f3c7155865
fix(engine): surface cancellation reasons in waits
2026-05-12 12:09:46 -05:00
Hunter Bown
1372d72627
fix(tools): unify tool-result retrieval references
2026-05-12 12:09:40 -05:00
reidliu41
a8554fdee3
fix(CI): validate pandoc output path before probing binary
...
Validate binary pandoc target formats before checking whether pandoc is
installed. This keeps invalid input errors stable on systems without pandoc
and fixes the CI test that expects missing output_path to be reported first.
2026-05-12 18:37:40 +08:00
masonxhuang
41a92b93f4
fix(tui): preserve prefix cache while pruning tool results
2026-05-12 17:49:41 +08:00
Duducoco
01044d652e
feat(init): update AGENTS.md in place instead of refusing
...
- overwrite existing AGENTS.md on /init rather than blocking with an "already exists" message
- report "Updated" vs "Created" depending on prior file state
- replace noop test with one verifying stale content is regenerated
2026-05-12 17:16:15 +08:00
kitty
fbab2beaeb
refactor(composer): extract magic values in derive_session_title to constants
2026-05-12 16:06:41 +08:00
kitty
e8b4f9911d
feat(composer): display session title in top-right of composer border
...
Show the current session's persisted metadata.title in the composer
border's top-right corner alongside the existing vim mode indicator.
- app.rs: add `session_title: Option<String>` field to App
- ui.rs: populate it from metadata.title in apply_loaded_session and
SessionUpdated handler; add derive_session_title() fallback helper
- widgets/mod.rs: render title (muted) + vim label in a single
right-aligned title_top span to avoid overlap
2026-05-12 15:58:06 +08:00
Hunter Bown
33822424d8
fix(npm): bump deepseekBinaryVersion to 0.8.32
2026-05-12 02:57:41 -05:00