feat(feedback): add structured feedback reporting

[elkaix]

Summary

• add structured /feedback support with feedback types, redacted session context, opt-in transcript/diff/tool details, and default send confirmation
• add compact GitHub fallback URLs and richer feedback-worker issue responses
• add tests for parsing, redaction, confirmation, fallback behavior, update/welcome banner behavior, and related update state handling
• keep existing branch changes to welcome/update UX and Dependabot configuration

Verification

• make check-pythinker-code
• make test-pythinker-code
• uv run pytest tests/ui_and_conv/test_shell_feedback_slash.py -q
• cd examples/feedback-worker && npm exec --package typescript@5.9.3 -- tsc --noEmit --strict --target ES2022 --module ESNext --moduleResolution Bundler src/index.ts
• git diff --check
• graphify update .

Summary by CodeRabbit

• New Features

  • Async feedback submission: structured, redacted session reports (bug/feature/ux/wrong) with prompts, optional diff/transcript/tool-details, confirmation, and GitHub fallback.
  • CLI welcome banner that surfaces update availability and one‑time “what’s new” post‑upgrade.

• Bug Fixes / Improvements

  • More robust feedback flow: JSON error handling, network timeout/retry fallbacks, clearer success messages.

• Tests

  • Expanded coverage for feedback parsing/submission, redaction, fallbacks, and welcome/update banner behaviors.

• Chores

  • Dependabot: weekly schedule, PR limits, grouping, and ignores for specified packages.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: a2560b84-90af-48d2-b8f1-8de7f06a54d3

📥 Commits

Reviewing files that changed from the base of the PR and between bc029cd and 1a648a3.

📒 Files selected for processing (4)

• src/pythinker_code/feedback.py
• src/pythinker_code/ui/shell/update.py
• tests/ui_and_conv/test_shell_feedback_slash.py
• tests/ui_and_conv/test_shell_update.py

---

📝 Walkthrough

Walkthrough

Adds a structured feedback submission flow with redaction and async backend submission (with GitHub fallback), integrates a welcome-banner update/what's-new chip, expands tests for both features, and tightens Dependabot configuration.

Changes

Feedback and Update Notification System

Layer / File(s)	Summary
Dependabot configuration: schedule, limits, and dependency ignores .github/dependabot.yml	Sets schedule to weekly (Mondays), caps open PRs to 3, groups minor/patch updates, and ignores ruff >=0.15 and click >=8.4.
Backend payload handling and sanitizers examples/feedback-worker/src/index.ts	Expands FeedbackPayload schema, adds JSON parse error handling, computes nested/top-level recent_errors, introduces MAX_CONTEXT_ITEMS, implements sanitizeValue/sanitizeRecord/sanitizeArray, appends JSON sections to GitHub body, returns {number, html_url}, and adds unique() helper.
Feedback types, limits, and contracts src/pythinker_code/feedback.py	Adds feedback kind aliases, secret/token regexes, truncation and size limits, and exported FeedbackOptions and FeedbackSubmission dataclasses.
Core feedback logic: parsing, redaction, payload building, and submission src/pythinker_code/feedback.py	Parses /feedback args, implements redact_text/redact_value, builds a structured payload with git/context snapshots and recursive redaction, and submits via aiohttp with tolerant JSON parsing returning a FeedbackSubmission.
Context collection: git snapshot, messages, tools, and subagents src/pythinker_code/feedback.py	Collects git branch/head/dirty/diff, recent visible messages (excluding ThinkPart), tool-call summaries and hints, and subagent meta.json records; supports truncation, permissive JSON parsing, and redaction.
Async /feedback slash command with prompts and submission src/pythinker_code/ui/shell/slash.py	Replaces the synchronous command with an async flow that parses args, prompts for message/confirmation (unless --yes), builds payload, submits async, falls back to creating/opening a GitHub issue on error, emits telemetry, and prints contextual success/fallback messages.
Update notification helpers and persisted marker src/pythinker_code/ui/shell/update.py	Adds LAST_SEEN_VERSION_FILE, concurrency-safe read/write, cached update candidate computation, welcome_update_target() for banner display, and consume_whats_new() for one-time post-upgrade notifications; refactors pending_update_notice().
Welcome banner UI: imports and integration with update/what's-new chip src/pythinker_code/ui/shell/__init__.py	Imports update helpers, adds _welcome_banner_chip() with precedence (update > what's-new), updates Shell.run to pass banner, and adjusts _print_welcome_info() layout to render the chip.
Tests and changelog updates tests/ui_and_conv/*, CHANGELOG.md	Adds/extends tests for /feedback fallback and submission paths, redaction and helpers, welcome/update helpers and banner rendering, and documents the new /feedback behavior in CHANGELOG.

🎯 4 (Complex) | ⏱️ ~45 minutes

Suggested labels

enhancement

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 14.12% which is insufficient. The required threshold is 70.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title follows conventional commits format with type 'feat' and scope 'feedback', clearly describing the main feature addition of structured feedback reporting.
Description check	✅ Passed	The description includes a summary of changes, verification steps, and addresses the template requirements with substantive detail about the feature implementation.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feature/structured-feedback

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 4

🤖 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/feedback.py`:
- Around line 148-154: The try/except around Path.home() is too broad and
silently swallows errors; replace the bare except with a narrow catch for the
specific exceptions that Path.home() can raise (e.g., RuntimeError and OSError)
and log the failure instead of passing silently — update the block that calls
Path.home() (and modifies the redacted variable) to except (RuntimeError,
OSError) as e: and call the module/logger (e.g., logger.warning or
processLogger) with a short message including the exception before continuing to
return redacted.

In `@src/pythinker_code/ui/shell/update.py`:
- Around line 387-391: The _read_last_seen_version function currently catches
all OSError and returns None, which hides real failures; change the exception
handling so that FileNotFoundError (or PermissionError due to a missing file)
returns None but other OSError cases are not silently swallowed—either re-raise
them or log and re-raise. Specifically, update the try/except around
LAST_SEEN_VERSION_FILE.read_text(...) so it catches FileNotFoundError and
returns None, and lets other OSError exceptions propagate (or log via the module
logger before raising) to preserve actionable errors.

In `@tests/ui_and_conv/test_shell_feedback_slash.py`:
- Around line 94-150: The test uses SimpleNamespace(...) as the mocked return
for submit_mock in test_submits_structured_payload; replace that with an actual
FeedbackSubmission instance to make the test robust to type changes: import
FeedbackSubmission from pythinker_code.feedback (or the module where it’s
defined), create FeedbackSubmission(number=42, html_url="https://issue/42") for
the AsyncMock return (the submit_mock used via
monkeypatch.setattr("pythinker_code.feedback.submit_feedback_payload",
submit_mock)), and keep the rest of the assertions the same so
submit_mock.assert_awaited_once() and payload inspection continue to work.

In `@tests/ui_and_conv/test_shell_update.py`:
- Around line 1010-1020: The test only checks the baseline file is non-empty;
instead assert it contains the actual recorded version so a wrong value can't
slip through: after calling update.consume_whats_new() in
test_consume_whats_new_baseline_on_first_launch, read
last_seen_file.read_text(...).strip() and assert it equals the expected version
exported by the module (e.g. update.__version__ or the module's canonical
version symbol), referencing update.consume_whats_new and
update.LAST_SEEN_VERSION_FILE to locate the logic to validate.

🪄 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: 05bb6b68-e18a-474f-ac54-f34d9ee3ea32

📥 Commits

Reviewing files that changed from the base of the PR and between 87bad36 and 22aaaf1.

📒 Files selected for processing (9)

• .github/dependabot.yml
• examples/feedback-worker/src/index.ts
• src/pythinker_code/feedback.py
• src/pythinker_code/ui/shell/__init__.py
• src/pythinker_code/ui/shell/slash.py
• src/pythinker_code/ui/shell/update.py
• tests/ui_and_conv/test_shell_feedback_slash.py
• tests/ui_and_conv/test_shell_update.py
• tests/ui_and_conv/test_shell_welcome_info.py

[elkaix]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.