feat(tui): TUI enhancements — adaptive theme, layout, and agent prompt overhaul - #102
Closed
elkaix wants to merge 18 commits into
Closed
feat(tui): TUI enhancements — adaptive theme, layout, and agent prompt overhaul#102elkaix wants to merge 18 commits into
elkaix wants to merge 18 commits into
Conversation
Turn recaps now default off; the agent only recaps when asked. Add a direct toggle — /config recaps on|off (also /settings recaps ...) — that persists to the config file and reloads the shell. The interactive /settings panel keeps its existing turn-recaps item. Part of the Codex TUI adoption backlog (item 0.1).
Port the Codex terminal-adaptation layer (codex-rs/tui) to Python: - ui/color_utils.py: hex parse/format, linear RGB blend, BT.601 luma. - ui/terminal_background.py: OSC 11 background probe (100ms timeout, per-process cache, PYTHINKER_NO_BG_PROBE opt-out) and theme = "auto" resolution with a dark fallback. /theme and the settings selector accept the new value. - terminal_capabilities.color_depth(): three usable color tiers (truecolor/256/16 + none) honoring FORCE_COLOR levels and the Windows Terminal WT_SESSION truecolor promotion. - get_diff_colors(): 16-color terminals now get plain green/red foreground diff styles instead of quantized hex background tints. /theme now compares against the persisted setting rather than the resolved active theme so users can pin dark/light while on auto. Backlog items 1.1-1.4.
Adopt the Codex renderer-safety behaviors: - Syntax-highlight size guard: code blocks beyond 512 KiB or 10k lines render as plain text with a 'highlighting skipped (N lines)' notice instead of paying an unbounded Pygments cost. - Large diff guard: expanded diffs are capped at 400 rendered lines (head + tail with an explicit omitted-line count) so one huge edit cannot freeze or flood the terminal. - Generic tool output switches from head-only to head-tail truncation, keeping the start (identifies the result) and the end (the actionable part) with an omitted-line notice. - Fence unwrapping for tables: ```md/```markdown fences whose body contains a header+delimiter table pair now render as markdown instead of opaque code. Conservative Codex heuristics: other languages, untagged fences, md fences without tables, and unclosed fences pass through unchanged. Backlog items 2.6, 3.1, 6.2, 7.5.
User-directed design wave on top of the Codex adoption Phase 1: - Transcript marker is now ⏺ (U+23FA) on macOS/Linux — Windows keeps the text circle, ASCII mode keeps the star. Tool/assistant rows blink the marker while running (reduced-motion pins it static) and settle to the solid green marker when finished; thinking rows carry the marker too. - Activity coral muted to a clay ramp (#C68D7E/#D8AC9E/#E9CDC2 dark, #B26A52/#9E563E/#82412D light). Shimmer simplified from wave+splash to a calm bidirectional sweep with settle beats; truecolor terminals get a continuous cosine-blended sheen via ui.color_utils.blend, lower tiers keep the discrete ramp. - Activity/todo headers share one metadata design: 'Verb… (12s, ↓ 2.4k tokens, 45 t/s)' — parenthesized, comma-separated, with a live tokens/sec readout on the working indicator and the pinned todo header (sliding-window rate over the turn's context tokens). - Thinking-effort frame colors form a cold→hot gradient ending on dark red for xhigh (slate→blue→teal→amber→orange→red). - Pinned todos: the active task title+box are coral; concurrent in-progress rows read light grey so the running task stays unmistakable. - Diff word-level highlights drop reverse-video for the theme's add/del highlight backgrounds (GitHub-style emphasis, no glare). - Turn recap is padded to the card inset instead of spanning edge-to-edge. Hardening (findings from the in-app review validated and fixed): - Terminal probe: catch select's ValueError (fd >= FD_SETSIZE), suppress tcsetattr restore failures, cap the OSC reply buffer at 4 KiB, and serialize the probe cache behind a lock. - Generic tool-output head/tail truncation halves the char budget per side so combined output can never exceed the limit. - /settings arg parsing splits the mode string once. - Added ~~~md tilde-fence unwrap coverage.
Second design wave driven by side-by-side comparison with the reference transcripts: - Tool headers use the parenthesized single-line form — ⏺ Bash(cmd…), ⏺ Update(path) — ellipsizing at the terminal edge instead of wrapping. - Result gutters no longer pad rows with trailing spaces to the terminal edge (copy-clean, ragged-right like the reference). - Todo list matches the reference: coral ■ box with bold default-color (white) title for in-progress rows, green ✓ with struck muted titles for done rows; coral stays on the top activity line. - Diff palette set to the specified values: row tints #052e05/#3a0808, sign/line-number accents #81C784/#E57373, content in the terminal's default (white) over the tint. Also fixes a span-layering bug where the row restyle buried word-level highlights. - Dark text hierarchy: primary output #D4D4D4; UI chrome (gutters, line numbers, expand hints, toolbar metadata) #6F6F6F. - Thinking-effort input bars dim their gradient color 30% toward the background pole so the frame hints without shouting. - The 'N background agents' line is gone everywhere — the bottom toolbar owns that count; the verb spinner + todos remain. Background subagent status rows hang-indent under their label. - Welcome banner gains a bold /init tip when the repo has no AGENTS.md or CLAUDE.md.
The pinned todo list is drawn by two code paths — the live view during a foreground turn and the prompt-side background status between turns — and they had drifted apart (◼/◻/✔ vs ■/□/✓ glyphs, hex off-white vs terminal-default bold titles, bright pending rows, no strikethrough). That made running tasks appear to change style mid-session. Both paths now render identically: coral ■ with a bold default-color (white) title for running rows, muted □ pending, green ✓ with struck muted titles when done.
… consistency
Three fixes from live-session screenshots:
- Pinned todo rows no longer carry a muted base style: the bold
running-task title sets no color of its own, so the muted base bled
through and rendered it grey instead of the terminal-default white.
The base style is gone; prefix/icon/title each carry their own style.
- The background working status line ('Ebbing…') now shows the same
'(elapsed, ↓ Nk tokens, N t/s)' metadata as the live working
indicator: elapsed since background work appeared and a 1.5s
sliding-window token rate over the status snapshot, dropped first on
narrow terminals and reset when background work drains.
- Word-level diff highlights are gated on line similarity (ratio >= 0.5):
heavy single-line rewrites previously flooded the whole row with the
brighter highlight tint, reading as a different palette from plain
added/removed rows. Such rewrites now render as plain rows.
Completes the c8c0f05 commit: the prompt.py change was clobbered from the index by a concurrent session's git operation before that commit landed, leaving the renderer half-committed. The background working status line ('Ebbing…') now carries the same '(elapsed, ↓ Nk tokens, N t/s)' metadata as the live working indicator: elapsed since background work first appeared and a 1.5s sliding-window token rate over the status snapshot's context tokens. The metadata is dropped first on narrow terminals and both trackers reset when background work drains.
Notification, progress-note, question-answered, and suggestion rows in the transcript still rendered with BulletColumns' default list bullet. They now carry the record marker in the block's accent color (severity for notifications, green for progress/answers, accent for suggestions). Genuine list rows — /help sections and nested subagent detail lines — deliberately keep the list dot.
Refactor all default agent YAML prompts with explicit Mission / Hard Constraints / Workflow / Output Contract sections for consistency and clarity. Update system.md overlay accordingly. Harden background manager, subagent runner, and agent tool to align with the Codex TUI adoption (Phase 1) gap map: stale-record reconciliation, resume contract enforcement, and interactive-visualizer fixes. Update all affected unit and e2e tests.
Contributor
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughOne cohesive PR: agent prompts rewritten; terminal/theme auto-detection and TUI renderer overhauled (markdown, diff, shimmer, markers, truncation); file/host/SSRF safety and background-agent resume reconciliation added; wire/session replay and compaction hardened; extensive tests updated. ChangesPrompt and policy updates
TUI, theme, and renderer
Background agents & orchestration
Host, file tools, and security
Runtime, web, wire, and locks
Sanitization & telemetry
Estimated code review effort Possibly related PRs
Suggested labels ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
|
Codecov Report❌ Patch coverage is 📢 Thoughts on this report? Let us know! |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 6
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
tests/core/test_agent_spec.py (1)
117-182: 🛠️ Refactor suggestion | 🟠 Major | 🏗️ Heavy liftReplace full prompt-body snapshots with invariant checks.
These blocks assert long prompt text verbatim, which makes the tests brittle to non-functional wording edits. Keep only contract-level invariants (required sections, key constraints, and tool policy assertions), and move away from full-body snapshots.
As per coding guidelines, “tests/**/*.py: Avoid brittle prompt tests that assert large prompt snapshots; prefer behavior, required sections, and exact small invariants.”
Also applies to: 254-299, 370-425, 496-515
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@tests/core/test_agent_spec.py` around lines 117 - 182, The test currently asserts the entire prompt body verbatim (see subagent_specs["coder"].system_prompt_args snapshot) which is brittle; update the tests to instead assert a small set of contract-level invariants: check that required section headers (e.g., "Mission", "Hard Constraints", "Workflow", "Output Contract") appear, that key constraints like "Stay tightly scoped" and "Never leave placeholders" are present, and that the artifact/structured-output requirement (the <coding_artifact> JSON fields) and tool/policy directives are enforced; replace the full-string snapshot assertions at the referenced blocks (around subagent_specs["coder"].system_prompt_args and the other noted ranges) with assertions using .contains() or regex matches for those specific phrases and that the snapshot only stores a minimal list of expected invariants instead of the whole prompt body.Source: Coding guidelines
src/pythinker_code/ui/shell/visualize/_live_view.py (1)
868-895:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winReset token-rate samples on turn boundaries.
_turn_token_rate()at Line 654 accumulates a sliding window, but top-level turn transitions (Lines 868-895) never clearself._turn_token_samples. A new turn started quickly after the previous one can show stale t/s from the prior turn.Suggested fix
@@ case TurnBegin(user_input=user_input): if self._active_turn_depth == 0: self._turn_start_time = time.monotonic() + self._turn_token_samples.clear() @@ case TurnEnd(): self._active_turn_depth = max(0, self._active_turn_depth - 1) if self._active_turn_depth == 0: self._turn_start_time = None + self._turn_token_samples.clear() self._pending_turn_recap = True @@ if is_interrupt: self._active_turn_depth = 0 self._turn_start_time = None + self._turn_token_samples.clear()Also applies to: 654-679, 217-220
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/pythinker_code/ui/shell/visualize/_live_view.py` around lines 868 - 895, The issue: _turn_token_rate() accumulates sliding-window samples in self._turn_token_samples but top-level turn transitions (handled in the TurnBegin and TurnEnd branches and in SteerInput/cleanup flows) never clear them, causing stale token/sec values to carry into new turns; fix by clearing/resetting self._turn_token_samples (and any related sampling state used by _turn_token_rate) whenever a top-level turn starts or fully ends—specifically, add code to clear self._turn_token_samples in the TurnBegin branch when self._active_turn_depth == 0 (where _turn_start_time, _recap_user_input, etc. are initialized) and in the TurnEnd branch when self._active_turn_depth becomes 0 (where _turn_start_time is set to None and _pending_turn_recap is set), and likewise ensure SteerInput/cleanup paths that begin an interrupting flow also reset self._turn_token_samples so samples never carry across independent turns.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/pythinker_code/app.py`:
- Around line 891-893: The AGENTS.md/CLAUDE.md existence check currently uses
Path.cwd() (work_dir variable) but must use the active session directory; change
the detection to check self._runtime.session.work_dir (or the existing session
work_dir variable) instead of Path.cwd(), e.g., compute session_dir =
self._runtime.session.work_dir and test (session_dir / "AGENTS.md").exists() ||
(session_dir / "CLAUDE.md").exists(); preserve the OSError handling and ensure
any subsequent logic that references work_dir uses the session_dir value so the
/init tip reflects the correct session path.
In `@src/pythinker_code/ui/color_utils.py`:
- Line 42: The docstring in src/pythinker_code/ui/color_utils.py contains an en
dash in "0–255" which triggers Ruff RUF002; edit the module/function docstring
(the line reading 'ITU-R BT.601 perceived brightness in the 0–255 range.') and
replace the en dash (–) with an ASCII hyphen (-) so it reads "0-255".
In `@src/pythinker_code/ui/shell/slash.py`:
- Around line 1295-1297: The warning string built in the f-string that uses
_t_set.warning is ambiguous; update the copy to reference the correct CLI flag
name so it matches other messages (use --config-file rather than --config), e.g.
change the message that currently reads "restart without --config text" to
"restart without --config-file text" (locate the f-string around _t_set.warning
in src/pythinker_code/ui/shell/slash.py and replace the token accordingly).
In `@src/pythinker_code/ui/terminal_background.py`:
- Around line 72-112: The _probe_uncached probe currently swallows all
exceptions and returns None; change its exception handlers to capture the
exception (e.g., except (ValueError, OSError) as exc) and emit a debug-level log
before returning so failures are visible while behavior remains the same.
Specifically, in _probe_uncached add logging in the initial stdin/stdout TTY
check except block, the termios.tcgetattr except block, and the outer probe
except (OSError, ValueError) block; include contextual info such as the fd (if
available), timeout/_OSC11_QUERY, and the exception instance, using the module
logger (logging.getLogger(__name__)) or the existing logger, then continue
returning None. Ensure the finally block still suppresses restore errors but
consider logging suppressed termios.tcsetattr failures at debug level as well;
do not change return values or control flow and keep parsing delegated to
parse_osc11_response.
In `@tests/ui_and_conv/test_output_guards.py`:
- Around line 21-24: Annotate the autouse fixture _isolated_registry with an
explicit generator/iterator return type to satisfy ANN202; change its signature
to something like def _isolated_registry() -> typing.Generator[None, None, None]
(or typing.Iterator[None]) and add the corresponding import from typing at the
top if missing, keeping the body that calls clear_tool_renderers() unchanged.
In `@tests/ui_and_conv/test_streaming_content_block.py`:
- Line 265: The current assertion only checks for any parentheses in output;
strengthen it by asserting the compact-metadata pattern is present: use a regex
search on the test's output variable to ensure there is a parenthesized metadata
block containing the "activity" key (e.g., match a pattern like
"(...activity...)" or a parenthesis-enclosed comma-separated key:value list)
rather than just any "(" or ")". Update the assertion in
tests/ui_and_conv/test_streaming_content_block.py that references output to use
re.search with an appropriate pattern (and add an import for re if missing) so
the test fails unless properly formatted compact-metadata is present.
---
Outside diff comments:
In `@src/pythinker_code/ui/shell/visualize/_live_view.py`:
- Around line 868-895: The issue: _turn_token_rate() accumulates sliding-window
samples in self._turn_token_samples but top-level turn transitions (handled in
the TurnBegin and TurnEnd branches and in SteerInput/cleanup flows) never clear
them, causing stale token/sec values to carry into new turns; fix by
clearing/resetting self._turn_token_samples (and any related sampling state used
by _turn_token_rate) whenever a top-level turn starts or fully
ends—specifically, add code to clear self._turn_token_samples in the TurnBegin
branch when self._active_turn_depth == 0 (where _turn_start_time,
_recap_user_input, etc. are initialized) and in the TurnEnd branch when
self._active_turn_depth becomes 0 (where _turn_start_time is set to None and
_pending_turn_recap is set), and likewise ensure SteerInput/cleanup paths that
begin an interrupting flow also reset self._turn_token_samples so samples never
carry across independent turns.
In `@tests/core/test_agent_spec.py`:
- Around line 117-182: The test currently asserts the entire prompt body
verbatim (see subagent_specs["coder"].system_prompt_args snapshot) which is
brittle; update the tests to instead assert a small set of contract-level
invariants: check that required section headers (e.g., "Mission", "Hard
Constraints", "Workflow", "Output Contract") appear, that key constraints like
"Stay tightly scoped" and "Never leave placeholders" are present, and that the
artifact/structured-output requirement (the <coding_artifact> JSON fields) and
tool/policy directives are enforced; replace the full-string snapshot assertions
at the referenced blocks (around subagent_specs["coder"].system_prompt_args and
the other noted ranges) with assertions using .contains() or regex matches for
those specific phrases and that the snapshot only stores a minimal list of
expected invariants instead of the whole prompt body.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 66f4c405-8357-4477-b33b-4c7164afc523
📒 Files selected for processing (73)
CHANGELOG.mdsrc/pythinker_code/agents/default/ask.yamlsrc/pythinker_code/agents/default/code_reviewer.yamlsrc/pythinker_code/agents/default/coder.yamlsrc/pythinker_code/agents/default/debug.yamlsrc/pythinker_code/agents/default/debugger.yamlsrc/pythinker_code/agents/default/explore.yamlsrc/pythinker_code/agents/default/implementer.yamlsrc/pythinker_code/agents/default/judge.yamlsrc/pythinker_code/agents/default/plan.yamlsrc/pythinker_code/agents/default/planner.yamlsrc/pythinker_code/agents/default/review.yamlsrc/pythinker_code/agents/default/scout.yamlsrc/pythinker_code/agents/default/security_reviewer.yamlsrc/pythinker_code/agents/default/system.mdsrc/pythinker_code/agents/default/verifier.yamlsrc/pythinker_code/app.pysrc/pythinker_code/background/manager.pysrc/pythinker_code/config.pysrc/pythinker_code/subagents/runner.pysrc/pythinker_code/tools/agent/__init__.pysrc/pythinker_code/ui/color_utils.pysrc/pythinker_code/ui/shell/__init__.pysrc/pythinker_code/ui/shell/components/diff.pysrc/pythinker_code/ui/shell/components/markdown.pysrc/pythinker_code/ui/shell/components/render_utils.pysrc/pythinker_code/ui/shell/components/tool_execution.pysrc/pythinker_code/ui/shell/glyphs.pysrc/pythinker_code/ui/shell/motion.pysrc/pythinker_code/ui/shell/prompt.pysrc/pythinker_code/ui/shell/render_constants.pysrc/pythinker_code/ui/shell/selectors/settings.pysrc/pythinker_code/ui/shell/slash.pysrc/pythinker_code/ui/shell/tool_renderers/_file_diff.pysrc/pythinker_code/ui/shell/tool_renderers/_render_utils.pysrc/pythinker_code/ui/shell/tool_renderers/agent.pysrc/pythinker_code/ui/shell/visualize/_blocks.pysrc/pythinker_code/ui/shell/visualize/_interactive.pysrc/pythinker_code/ui/shell/visualize/_live_view.pysrc/pythinker_code/ui/terminal_background.pysrc/pythinker_code/ui/terminal_capabilities.pysrc/pythinker_code/ui/theme.pysrc/pythinker_code/utils/rich/columns.pytasks/todo.mdtests/background/test_manager.pytests/core/test_agent_spec.pytests/core/test_config.pytests/core/test_default_agent.pytests/e2e/test_shell_modal_e2e.pytests/e2e/test_shell_pty_e2e.pytests/tools/test_agent_tool.pytests/ui/test_shell_markdown.pytests/ui_and_conv/test_color_utils.pytests/ui_and_conv/test_empty_think_part_indicator.pytests/ui_and_conv/test_live_view_notifications.pytests/ui_and_conv/test_live_view_todos.pytests/ui_and_conv/test_markdown_guards.pytests/ui_and_conv/test_modal_lifecycle.pytests/ui_and_conv/test_output_guards.pytests/ui_and_conv/test_prompt_tips.pytests/ui_and_conv/test_settings_recaps_slash.pytests/ui_and_conv/test_shell_motion.pytests/ui_and_conv/test_shell_motion_shimmer.pytests/ui_and_conv/test_streaming_content_block.pytests/ui_and_conv/test_terminal_background.pytests/ui_and_conv/test_terminal_capabilities.pytests/ui_and_conv/test_theme.pytests/ui_and_conv/test_thinking_cycle.pytests/ui_and_conv/test_tui_blocks_integration.pytests/ui_and_conv/test_tui_card_tool_renderers.pytests/ui_and_conv/test_tui_components.pytests/ui_and_conv/test_tui_theme_tokens.pytests/ui_and_conv/test_visualize_running_prompt.py
Implements the remediation plan for the validated audit findings (1 Critical, 15 High, 29 Medium, 20 Low) across five phases, each fix behind a TDD test and gated by `make check` plus a security review. Phase 0 — shell permission classifier: close the read-only / auto-mode bypass cluster (interior &/|& separators, casefolded base commands, wrapper value-options, find/xargs/awk payloads, glued output redirection, unsafe `git -c`, uv sub-namespaces and `uv run` option-prefix bypass). Phase 1 — confinement, egress, telemetry: symlink-resolve file read/write/edit and grep before workspace/sensitive checks; fail-closed SSRF with a connection-pinned resolver; bounded shell wait; Sentry path/home redaction; invisible-char and case-folded sensitive-file handling; sensitive-import gate; untrusted-output wrapping; subagent-id path validation. Phase 2 — tool dispatch, lifecycle, context integrity: MCP-vs-builtin tool collisions; run_soul task-leak cleanup; restore-id/path traversal guards; compaction rollback; mid-tool-cancel turn balance; cyclic-extend detection. Phase 3 — wire server, auth, web-server: wire read-loop hardening; OAuth refresh / device-id / 403 handling; provider base_url validation; replay watermark; session-leak cleanup; ZIP-import validation. Phase 4 — UI/usage/CLI: ANSI sanitization at render boundaries (incl. generic tool arg-key names); usage-meter consumed-vs-remaining; reset-window loop guard; RunAgents approval fingerprint over child prompts; owner-only MCP config; live Typer help; /restore traversal and error handling; bounded approval-request store. Review-found gaps were fixed with regression tests: uv-run option bypass, grep symlink escape, SSH-key import gate, ANSI arg-key injection, and the MCP-config / share-dir permission race. Also hardens auth JSON parsing.
- app.py: detect AGENTS.md/CLAUDE.md in the session work_dir (awaiting the async HostPath.exists) instead of process cwd, so the /init tip is correct when cwd differs from the session path. - color_utils.py: replace EN DASH with ASCII hyphen in the luma docstring. - slash.py: clarify the recaps config-flag guidance (--config vs --config-file). - terminal_background.py: debug-log swallowed OSC11 probe failures instead of silently degrading, per the exception-handling rule. - tests: annotate the autouse fixture return type; tighten the thinking-status metadata assertion to validate the token-count block, not just any paren.
Qwen3.x/3.7 are hybrid thinking models: reasoning is toggleable per request
via the standard Anthropic thinking block, which the OpenCode Go
@ai-sdk/anthropic route (and Alibaba Model Studio's Anthropic-compatible
endpoint) accepts as {"type": "enabled", "budget_tokens": N} /
{"type": "disabled"}.
Previously OpenCode Go Qwen models got no thinking capability, so their
effort was uncontrollable — while the Alibaba plan already exposed it. Give
the Qwen family the controllable 'thinking' capability so create_llm routes
the selected effort through with_thinking, which (for these non-Claude
models) emits the budget-based payload and clamps xhigh/max -> high,
minimal -> low. GLM/MiniMax remain always_thinking (effort can't be turned
off); Qwen can.
Adds a clamp regression case pinning the budget-safe mapping for qwen3.7-max
so the budgets[...] lookup can never KeyError.
Contributor
There was a problem hiding this comment.
Actionable comments posted: 14
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/pythinker_code/cli/vis.py`:
- Around line 15-19: The parameter named ctx in the CLI command signature (the
typer.Context parameter in vis.py) is unused and triggers Ruff ARG001; rename it
to _ctx (or remove the parameter entirely if not required) in the function
signature where ctx: typer.Context is declared so the linter recognizes it as
intentionally unused; ensure any internal references (if any) are updated or
removed accordingly and keep the Annotated option for host unchanged.
In `@src/pythinker_code/file_restore.py`:
- Around line 83-97: The code currently decodes point.content_b64 with
base64.b64decode(point.content_b64 or "") which can silently write empty or
malformed data; update the restore flow in the block that follows
FileRestorePoint parsing (around restore_path, point, and before
point.path.write_bytes) to treat missing/empty or invalid base64 as a corrupt
restore point: first check that point.content_b64 is a non-empty string and if
not raise FileNotFoundError(f"Corrupt restore point: {restore_id}"), then
attempt to base64.b64decode(point.content_b64) inside a try/except that catches
base64/binascii decode errors (e.g., binascii.Error or ValueError) and re-raises
FileNotFoundError from the caught exception; only call
point.path.write_bytes(decoded_bytes) after a successful decode.
In `@src/pythinker_code/scratchpad.py`:
- Around line 438-439: The code currently deletes the lock file by calling
lock_file.unlink() inside a contextlib.suppress(OSError) after releasing the
file lock; remove that unlink so the lock file remains on disk and only the file
lock is released (e.g., via closing or releasing the flock handle). Update the
release logic in the function/method that manipulates the lock (references to
lock_file and contextlib.suppress) to stop unlinking and ensure the lock is
released correctly without deleting the *.scratchpad.lock inode.
In `@src/pythinker_code/soul/__init__.py`:
- Around line 240-244: The cleanup loop currently cancels then unconditionally
awaits each task (soul_task, cancel_event_task, notification_task), which can
re-raise a previously-failed exception and abort remaining shutdown steps
(wire.shutdown(), wire.join(), _current_wire.reset). Fix by skipping awaiting a
task that has already completed with an exception: after task.cancel(), check
task.done() and task.exception(), and only await the task (inside
contextlib.suppress(asyncio.CancelledError)) if it is not already done with a
non-None exception; this prevents re-raising past failures while still awaiting
normally-cancelled or successful tasks so shutdown (wire.shutdown(),
wire.join(), _current_wire.reset) always runs.
In `@src/pythinker_code/soul/permission.py`:
- Around line 214-217: The current sudo-unwrapping treats only short forms in
_SUDO_VALUE_OPTS as value-taking, allowing long options (e.g. --user) and
--opt=value to be misinterpreted as commands; update _SUDO_VALUE_OPTS to include
the long-form option names you support and change the unwrapping logic (the code
around the sudo unwrap block that iterates args at ~lines 551-567) to: treat any
token starting with "--" that exactly matches a long option in _SUDO_VALUE_OPTS
as consuming the next token, and treat tokens of the form "--opt=value" as also
consuming their value (i.e., do not treat the portion after '=' as a command);
also ensure the same pattern is applied consistently to the time-related logic
using _TIME_VALUE_OPTS.
- Around line 705-723: The function _uv_run_payload currently assumes the first
non-option token is "run" and misses cases with global options (e.g., "uv
--directory repo run ..."); change the logic to first scan args from the start
to skip any global options (respecting "--" and value-taking global options
using the same _UV_RUN_VALUE_OPTS set) until you encounter the literal "run"
token, return None if none found, then set rest = args[index_of_run + 1 :] and
continue the existing run-specific option skipping loop (using
_UV_RUN_VALUE_OPTS for run options) to extract and return the wrapped payload
tokens.
In `@src/pythinker_code/soul/pythinkersoul.py`:
- Around line 1668-1683: When cancellation happens in the try around
result.tool_results(), we must not overwrite already-streamed tool outputs;
change the except asyncio.CancelledError branch to collect the set of completed
results (the ToolResult instances that were returned before cancellation or that
were already emitted via on_tool_result) by inspecting whatever
partial/available results you can get from result (e.g., the portion of
result.tool_results() that succeeded or a record of IDs emitted by
on_tool_result), then synthesize ToolResult(tool_call_id=...,
return_value=ToolRuntimeError(...)) only for tool_call IDs present in
result.tool_calls but missing from the completed set, combine completed results
+ synthesized interrupted results and pass that combined list into
self._grow_context(result, interrupted_or_combined) via asyncio.shield, and
finally re-raise the CancelledError; reference symbols: result.tool_results(),
on_tool_result, ToolResult, ToolRuntimeError, self._grow_context,
result.tool_calls.
In `@src/pythinker_code/soul/slash.py`:
- Around line 325-327: The current parsing uses args.split() which breaks
quoted/escaped paths and mangles targets; replace that with a proper shell-style
parser (e.g., use shlex.split(args) or an argparse.ArgumentParser) to correctly
handle quoting/escaping and to parse the --force flag and the target path;
update the code in slash.py (replace args.split() usage where tokens =
args.split(), force = "--force" in tokens, target = sanitize_cli_path(...)) to
parse with shlex.split(args) or an ArgumentParser and derive force (bool) and
target (the remaining positional argument, sanitized with sanitize_cli_path),
and remove/replace the duplicated parsing logic in
src/pythinker_code/ui/shell/export_import.py by reusing the same parsing
approach or centralizing it into a shared helper (e.g., a parse_import_args
function) so both modules behave identically.
In `@src/pythinker_code/tools/web/fetch.py`:
- Around line 28-33: The current _ip_is_blocked function returns False on
ipaddress.ip_address(address) ValueError which allows unknown/invalid addresses
to proceed; update the except ValueError branch in _ip_is_blocked to return True
(block) instead, so any parse error results in a blocked outcome; locate the
function _ip_is_blocked and change the exception handler for
ipaddress.ip_address to return True while keeping the final return of (not
ip.is_global) or ip.is_multicast.
In `@src/pythinker_code/web/runner/process.py`:
- Around line 570-592: The async function add_websocket_and_begin_replay
performs a blocking filesystem stat via wire_file.stat().st_size; replace that
call with an async stat (e.g., anyio.Path or aiofiles) so the event loop isn't
blocked: convert wire_file to an async path (for example anyio.Path(wire_file)),
await its .stat() and read .st_size, keep the same OSError handling, and ensure
you import the chosen async filesystem API; update the watermark assignment
inside add_websocket_and_begin_replay accordingly so it awaits the async stat
rather than calling the blocking Path.stat().
- Around line 676-683: The current broad except Exception in the error recovery
block (inside the handler that checks isinstance(in_message,
JSONRPCPromptMessage) and manipulates self._in_flight_prompt_ids, was_busy, and
await self._emit_status) should be narrowed to only the intended error classes
to avoid masking bugs; change the handler to catch a tuple of expected errors
(e.g., ValueError, OSError, and json.decoder.JSONDecodeError) or split into
separate except blocks for validation errors versus I/O errors, keeping the same
cleanup logic (discarding in_message.id, checking was_busy and calling
_emit_status("idle", reason="prompt_error")) and still logging the exception
class and message via logger.error.
In `@tests/core/test_project_memory.py`:
- Around line 254-256: The local async helper slow_write is missing a return
type annotation; update its signature to include an explicit return type (e.g.,
async def slow_write(self, target, entries) -> None:) so the coroutine's return
type is declared, keeping the body calling await orig_write(self, target,
entries) unchanged and referencing the existing orig_write symbol.
In `@tests/e2e/test_cli_error_output.py`:
- Line 207: The subprocess.run call in tests/e2e/test_cli_error_output.py should
explicitly pass check=False to satisfy the linter and clarify intent; update the
return subprocess.run(...) invocation used in the test to include check=False
(since the test inspects result.returncode rather than raising exceptions) so
Ruff no longer warns about a missing check argument.
In `@tests/ui_and_conv/test_export_import.py`:
- Around line 1402-1411: Add assertions that pin the token-count side effect
when testing perform_import: in the non-force case (where you currently assert
isinstance(result, str) and ctx.append_message.assert_not_awaited()), also
assert ctx.update_token_count.assert_not_awaited() to ensure token counts aren’t
changed; in the force=True case (where you assert isinstance(result2, tuple) and
ctx.append_message.assert_awaited_once()), also assert
ctx.update_token_count.assert_awaited_once(). Apply the same pair of assertions
to the other identical test block mentioned (the second import test around the
later assertions) so both refused and forced import paths verify
update_token_count usage alongside append_message.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: ee8e8376-0578-4f25-9dac-2ab27712d3ea
📒 Files selected for processing (127)
packages/pythinker-host/src/pythinker_host/__init__.pypackages/pythinker-host/src/pythinker_host/local.pypackages/pythinker-host/src/pythinker_host/path.pypackages/pythinker-host/src/pythinker_host/ssh.pypackages/pythinker-review/tests/unit/test_security_intel.pysrc/pythinker_code/__main__.pysrc/pythinker_code/acp/host.pysrc/pythinker_code/agentspec.pysrc/pythinker_code/app.pysrc/pythinker_code/approval_runtime/runtime.pysrc/pythinker_code/auth/lm_studio.pysrc/pythinker_code/auth/oauth.pysrc/pythinker_code/auth/ollama.pysrc/pythinker_code/auth/openai.pysrc/pythinker_code/cli/export.pysrc/pythinker_code/cli/mcp.pysrc/pythinker_code/cli/vis.pysrc/pythinker_code/cli/web.pysrc/pythinker_code/file_restore.pysrc/pythinker_code/hooks/engine.pysrc/pythinker_code/memory/recall.pysrc/pythinker_code/plugin/manager.pysrc/pythinker_code/project_memory.pysrc/pythinker_code/scratchpad.pysrc/pythinker_code/session_fork.pysrc/pythinker_code/share.pysrc/pythinker_code/soul/__init__.pysrc/pythinker_code/soul/compaction_restore.pysrc/pythinker_code/soul/permission.pysrc/pythinker_code/soul/pythinkersoul.pysrc/pythinker_code/soul/slash.pysrc/pythinker_code/soul/toolset.pysrc/pythinker_code/subagents/discovery.pysrc/pythinker_code/subagents/store.pysrc/pythinker_code/telemetry/errors.pysrc/pythinker_code/telemetry/sentry.pysrc/pythinker_code/tools/agent/__init__.pysrc/pythinker_code/tools/background/__init__.pysrc/pythinker_code/tools/file/grep_local.pysrc/pythinker_code/tools/file/read.pysrc/pythinker_code/tools/file/replace.pysrc/pythinker_code/tools/file/write.pysrc/pythinker_code/tools/shell/__init__.pysrc/pythinker_code/tools/web/fetch.pysrc/pythinker_code/ui/color_utils.pysrc/pythinker_code/ui/shell/components/bash_execution.pysrc/pythinker_code/ui/shell/export_import.pysrc/pythinker_code/ui/shell/slash.pysrc/pythinker_code/ui/shell/tool_renderers/_render_utils.pysrc/pythinker_code/ui/shell/usage.pysrc/pythinker_code/ui/shell/usage_adapters/alibaba.pysrc/pythinker_code/ui/shell/usage_adapters/minimax.pysrc/pythinker_code/ui/shell/usage_adapters/openai_chatgpt.pysrc/pythinker_code/ui/shell/visualize/_blocks.pysrc/pythinker_code/ui/shell/visualize/_worklog.pysrc/pythinker_code/ui/terminal_background.pysrc/pythinker_code/utils/aiohttp.pysrc/pythinker_code/utils/export.pysrc/pythinker_code/utils/path.pysrc/pythinker_code/utils/sensitive.pysrc/pythinker_code/utils/trust.pysrc/pythinker_code/vis/api/sessions.pysrc/pythinker_code/web/api/config.pysrc/pythinker_code/web/api/sessions.pysrc/pythinker_code/web/runner/process.pysrc/pythinker_code/wire/file.pysrc/pythinker_code/wire/server.pytests/auth/test_lm_studio_auth.pytests/auth/test_oauth_device_id.pytests/auth/test_oauth_refresh.pytests/auth/test_ollama_auth.pytests/auth/test_openai_auth.pytests/conftest.pytests/core/test_agent_spec.pytests/core/test_approval_auto.pytests/core/test_approval_runtime.pytests/core/test_cli_reload.pytests/core/test_compaction_restore.pytests/core/test_default_agent.pytests/core/test_export_cli.pytests/core/test_file_restore_points.pytests/core/test_mcp_docker_rm.pytests/core/test_notifications.pytests/core/test_permission_profiles.pytests/core/test_project_memory.pytests/core/test_pythinkersoul_turn_balance.pytests/core/test_recall_provider.pytests/core/test_scratchpad.pytests/core/test_session_fork.pytests/core/test_soul_import_command.pytests/core/test_startup_imports.pytests/core/test_subagent_discovery.pytests/core/test_subagent_store.pytests/core/test_toolset.pytests/core/test_wire_file_compat.pytests/core/test_wire_plan_mode.pytests/core/test_wire_server_steer.pytests/e2e/test_cli_error_output.pytests/hooks/test_engine.pytests/telemetry/test_sentry_filters.pytests/tools/test_agent_tool.pytests/tools/test_background_tools.pytests/tools/test_fetch_url.pytests/tools/test_grep.pytests/tools/test_read_file.pytests/tools/test_shell_bash.pytests/tools/test_smart_search.pytests/tools/test_str_replace_file.pytests/tools/test_untrusted_wrapping.pytests/tools/test_write_file.pytests/ui/usage_adapters/test_alibaba_adapter.pytests/ui/usage_adapters/test_minimax.pytests/ui/usage_adapters/test_openai_chatgpt.pytests/ui_and_conv/test_export_import.pytests/ui_and_conv/test_output_guards.pytests/ui_and_conv/test_shell_export_import_commands.pytests/ui_and_conv/test_shell_switch_slash.pytests/ui_and_conv/test_streaming_content_block.pytests/ui_and_conv/test_tui_card_tool_renderers.pytests/ui_and_conv/test_tui_components.pytests/ui_and_conv/test_visualize_running_prompt.pytests/ui_and_conv/test_worklog_render.pytests/utils/test_sensitive.pytests/utils/test_trust.pytests/vis/test_app.pytests/web/test_config_api_redaction.pytests/web/test_session_error_recovery.py
💤 Files with no reviewable changes (2)
- src/pythinker_code/plugin/manager.py
- src/pythinker_code/main.py
Stop recoloring the whole input border by thinking effort. The border is now one static frame grey at every level; the effort signal moves to a small label flushed right on the input's top border — a level-colored dot (off->max: slate->blue->teal->amber->orange->red) plus the muted level word. The dot carries the cold->hot color at full strength (it's a single glyph), while the word uses a muted class so it never competes with the typed text. The label is hidden for native-thinking models (always_thinking with no user dial) and non-thinking models, and the rule auto-shortens by the measured label width so the top line never wraps. _prompt_separator_style no longer borrows thinking_frame_style, so all input separators (top border and footer) render the same static frame grey.
…king Two related thinking-effort changes: 1. Remove the duplicated effort indicator from the footer (the 'agent <model> • <effort>' / 'native reasoning' text under the input). Effort now shows in exactly one place — the top-border label added earlier. The footer mode line is just 'agent <model>', degrading to the bare mode on narrow terminals. Drops the now-orphaned _thinking_footer_label helper. 2. Mark all Qwen models (qwen3.7-max/plus, qwen3.6-plus/flash, qwen3 coder plus/flash on Alibaba; qwen* on OpenCode Go) as always_thinking instead of the controllable 'thinking' capability. Qwen now matches GLM/MiniMax: reasoning is native and always on, with no user effort dial and no top-border effort label. Reasoning still flows over the Anthropic thinking block both Anthropic-compatible routes accept.
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/pythinker_code/ui/shell/prompt.py`:
- Around line 2478-2484: The top-border assembly currently always appends the
label which can cause wrapping; change the return logic to check if label_width
+ gap <= len(rule) (using the existing gap, label_width, rule_width, label,
border_style, and rule variables) and only include the label when it fits;
otherwise return just the plain rule segment (border_style with "─" * max(0,
len(rule) - gap - label_width) or simply border_style with "─" * len(rule) as
the fallback) so narrow terminals won’t wrap the label.
In `@tests/ui_and_conv/test_prompt_tips.py`:
- Line 608: The test function
test_card_toolbar_separator_is_static_grey_regardless_of_effort currently types
the monkeypatch fixture as Any; change its parameter annotation to the concrete
pytest.MonkeyPatch type (i.e., monkeypatch: pytest.MonkeyPatch) and ensure
pytest is imported in the test module so the annotation resolves, replacing the
Any usage to satisfy the ANN401 lint rule.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: d5547b6c-a99f-43e2-a62f-413e201c7b66
📒 Files selected for processing (7)
CHANGELOG.mdpackages/pythinker-core/tests/test_anthropic_thinking.pysrc/pythinker_code/auth/opencode_go.pysrc/pythinker_code/ui/shell/prompt.pysrc/pythinker_code/ui/theme.pytests/auth/test_opencode_go_auth.pytests/ui_and_conv/test_prompt_tips.py
Critical permission-classifier bypasses: - sudo long value-options (`sudo --user alice rm -rf /`) were not consumed, so the wrapped destructive payload classified as the option's value. Add the long forms to _SUDO_VALUE_OPTS. - uv global options before `run` (`uv --directory repo run rm -rf /`) hid the subcommand; the global flag's value was mistaken for it. Add _uv_strip_global_opts and apply it in both mutation and destructive paths. Major: - file_restore: treat missing (None) or malformed-base64 content as a corrupt restore point instead of silently writing an empty/garbage file. - web/fetch: _ip_is_blocked now fails closed (blocks) on an unparseable address. - scratchpad: stop unlinking the advisory lock file (split-inode race); keep it persistent and add *.scratchpad.lock to the written .gitignore patterns. - soul shutdown: don't re-await an already-finished task in the cleanup loop — retrieve its exception without re-raising so the rest of shutdown still runs. - pythinkersoul: on mid-tool interruption, keep the real results of calls that already completed (captured via on_tool_result) and only synthesize the interruption marker for still-pending calls. Minor / nitpick: - /import arg parsing now uses shlex via a shared parse_import_args helper (soul/slash + ui/shell/export_import), preserving quoted paths. - web/runner: offload the blocking wire-file stat with asyncio.to_thread; document the intentional broad except at the per-message dispatch boundary. - cli/vis: rename unused callback param to _ctx. - tests: regression cases for both bypasses; strengthened import token-count assertions; lock-file persistence test; minor annotations.
Contributor
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
src/pythinker_code/soul/pythinkersoul.py (1)
1681-1695:⚠️ Potential issue | 🟠 MajorKeep and drain the interrupted
_grow_contexttask before re-raising CancelledError.In the
except asyncio.CancelledErrorforresult.tool_results()(src/pythinker_code/soul/pythirersoul.py~1681-1696),asyncio.shield(asyncio.create_task(...))is used without storing the task reference; if cancellation hits again while awaiting the shielded wait, you can end up re-raising without having “drained” the same task like the normal path immediately below does (~1704-1710). Mirror that normal pattern (storegrow_context_taskand re-shield-await it onCancelledError) so the interrupted context write completes cleanly.Suggested fix
- await asyncio.shield(asyncio.create_task(self._grow_context(result, interrupted))) + grow_context_task = asyncio.create_task( + self._grow_context(result, interrupted) + ) + try: + await asyncio.shield(grow_context_task) + except asyncio.CancelledError: + await asyncio.shield(grow_context_task) raise🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/pythinker_code/soul/pythinkersoul.py` around lines 1681 - 1695, The current CancelledError handling creates a shielded task for self._grow_context(...) but doesn't keep a reference, so if another cancellation occurs while awaiting it the background grow_context write may never be drained; change the pattern to store the task in a variable (e.g., grow_context_task = asyncio.create_task(self._grow_context(result, interrupted))) and then await it via await asyncio.shield(grow_context_task) inside the except asyncio.CancelledError block (and re-shield-await again before re-raising) so the interrupted context write started by _grow_context (and the ToolResult/ToolRuntimeError entries built from result.tool_calls and completed_tool_results) is always completed/drained before letting CancelledError propagate.tests/e2e/test_cli_error_output.py (1)
199-216:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winTest helper uses a fixed path in
$HOMEinstead oftmp_path.
_run_pythinker_mainhardcodesPath.home() / ".pythinker-test-share"which leaves artifacts in the user's home directory and can cause flaky behavior under parallel test runs. Consider accepting atmp_pathfixture like_run_pythinkerdoes.Suggested fix
-def _run_pythinker_main(args: list[str]) -> subprocess.CompletedProcess[str]: +def _run_pythinker_main(args: list[str], *, share_dir: Path) -> subprocess.CompletedProcess[str]: """Run via python -m pythinker_code (__main__.py) rather than pythinker_code.cli.""" env = os.environ.copy() - env["PYTHINKER_SHARE_DIR"] = str(Path.home() / ".pythinker-test-share") + env["PYTHINKER_SHARE_DIR"] = str(share_dir) env["NO_COLOR"] = "1" ...Then update the test:
-def test_root_help_lists_live_options() -> None: +def test_root_help_lists_live_options(tmp_path: Path) -> None: ... - result = _run_pythinker_main(["--help"]) + result = _run_pythinker_main(["--help"], share_dir=tmp_path / "share")🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@tests/e2e/test_cli_error_output.py` around lines 199 - 216, The helper _run_pythinker_main currently hardcodes PYTHINKER_SHARE_DIR to Path.home() / ".pythinker-test-share"; change it to accept a tmp_path fixture (e.g., add parameter tmp_path: Path) and set env["PYTHINKER_SHARE_DIR"] = str(tmp_path / ".pythinker-test-share") so tests use an isolated temporary directory; update any calls to _run_pythinker_main (or prefer reusing the existing _run_pythinker helper) in the test file to pass the tmp_path fixture and remove reliance on the user HOME to avoid cross-test interference.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@src/pythinker_code/soul/pythinkersoul.py`:
- Around line 1681-1695: The current CancelledError handling creates a shielded
task for self._grow_context(...) but doesn't keep a reference, so if another
cancellation occurs while awaiting it the background grow_context write may
never be drained; change the pattern to store the task in a variable (e.g.,
grow_context_task = asyncio.create_task(self._grow_context(result,
interrupted))) and then await it via await asyncio.shield(grow_context_task)
inside the except asyncio.CancelledError block (and re-shield-await again before
re-raising) so the interrupted context write started by _grow_context (and the
ToolResult/ToolRuntimeError entries built from result.tool_calls and
completed_tool_results) is always completed/drained before letting
CancelledError propagate.
In `@tests/e2e/test_cli_error_output.py`:
- Around line 199-216: The helper _run_pythinker_main currently hardcodes
PYTHINKER_SHARE_DIR to Path.home() / ".pythinker-test-share"; change it to
accept a tmp_path fixture (e.g., add parameter tmp_path: Path) and set
env["PYTHINKER_SHARE_DIR"] = str(tmp_path / ".pythinker-test-share") so tests
use an isolated temporary directory; update any calls to _run_pythinker_main (or
prefer reusing the existing _run_pythinker helper) in the test file to pass the
tmp_path fixture and remove reliance on the user HOME to avoid cross-test
interference.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 454217a2-23d9-474c-8f24-605eb5b7a2bf
📒 Files selected for processing (16)
src/pythinker_code/cli/vis.pysrc/pythinker_code/file_restore.pysrc/pythinker_code/scratchpad.pysrc/pythinker_code/soul/__init__.pysrc/pythinker_code/soul/permission.pysrc/pythinker_code/soul/pythinkersoul.pysrc/pythinker_code/soul/slash.pysrc/pythinker_code/tools/web/fetch.pysrc/pythinker_code/ui/shell/export_import.pysrc/pythinker_code/utils/export.pysrc/pythinker_code/web/runner/process.pytests/core/test_permission_profiles.pytests/core/test_project_memory.pytests/core/test_scratchpad.pytests/e2e/test_cli_error_output.pytests/ui_and_conv/test_export_import.py
Member
Author
|
Superseded by #103 (branch renamed to feat/tui-enhancements-and-hardening, which contains this work plus the security-audit remediation, multi-instance hardening, and openai package refactor). |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Test plan
pytestpytest tests/core/test_agent_spec.py tests/core/test_default_agent.pypytest tests/e2e/Summary by CodeRabbit
New Features
Improvements
Stability & Security
Tests