docs: merge notification system docs into single Native-First guide

- Consolidate 5 notification-system-* files into doc/notification-system.md
- Add web-push cleanup guide and Start-on-Login glossary entry
- Configure markdownlint for consistent formatting
- Remove web-push references, focus on native OS scheduling

Reduces maintenance overhead while preserving all essential information
in a single, well-formatted reference document.

.cursor/rules/core/base_context.mdc

@@ -21,7 +21,7 @@ alwaysApply: false
 
 ## Purpose
 
-All interactions must *increase the human's competence over time* while
+All interactions must _increase the human's competence over time_ while
 completing the task efficiently. The model may handle menial work and memory
 extension, but must also promote learning, autonomy, and healthy work habits.
 The model should also **encourage human interaction and collaboration** rather
@@ -31,7 +31,7 @@ machine-driven steps.
 
 ## Principles
 
-1. Competence over convenience: finish the task *and* leave the human more
+1. Competence over convenience: finish the task _and_ leave the human more
 
    capable next time.
 
@@ -75,7 +75,7 @@ assumptions if unanswered.
 
 ### timebox_minutes
 
-*integer or null* — When set to a positive integer (e.g., `5`), this acts
+_integer or null_ — When set to a positive integer (e.g., `5`), this acts
 as a **time budget** guiding the model to prioritize delivering the most
 essential parts of the task within that constraint.
 
@@ -91,7 +91,7 @@ Behavior when set:
 
 3. **Signal Skipped Depth** — Omitted details should be listed under
 
-   *Deferred for depth*.
+   _Deferred for depth_.
 
 4. **Order by Value** — Start with blocking or high-value items, then
 
@@ -198,7 +198,7 @@ Default: Doer + short Mentor notes.
 
 ## Self-Check (model, before responding)
 
-- [ ] Task done *and* at least one competence lever included (≤120 words
+- [ ] Task done _and_ at least one competence lever included (≤120 words
   total)
 - [ ] At least one collaboration/discussion hook present
 - [ ] Output follows the **Output Contract** sections

.cursor/rules/core/harbor_pilot_universal.mdc

@@ -53,7 +53,7 @@ evidence-backed steps**.
 - **Verifiable Outputs**: Include expected results, status codes, or
   error messages
 
-- **Cite evidence** for *Works/Doesn't* items (timestamps, filenames,
+- **Cite evidence** for _Works/Doesn't_ items (timestamps, filenames,
   line numbers, IDs/status codes, or logs).
 
 ## Required Sections
@@ -181,8 +181,8 @@ Before publishing, verify:
 
 ---
 
-**Status**: 🚢 ACTIVE — General ruleset extending *Base Context — Human
-Competence First*
+**Status**: 🚢 ACTIVE — General ruleset extending _Base Context — Human
+Competence First_
 
 **Priority**: Critical
 **Estimated Effort**: Ongoing reference

.cursor/rules/core/less_complex.mdc

@@ -16,7 +16,7 @@ language: Match repository languages and conventions
    where it occurs; avoid new layers, indirection, or patterns unless
    strictly necessary.
 2. **Keep scope tight.** Implement only what is needed to satisfy the
-   acceptance criteria and tests for *this* issue.
+   acceptance criteria and tests for _this_ issue.
 3. **Avoid speculative abstractions.** Use the **Rule of Three**:
    don't extract helpers/patterns until the third concrete usage proves
    the shape.
@@ -29,7 +29,7 @@ language: Match repository languages and conventions
 7. **Targeted tests only.** Add the smallest set of tests that prove
    the fix and guard against regression; don't rewrite suites.
 8. **Document the "why enough."** Include a one-paragraph note
-   explaining why this minimal solution is sufficient *now*.
+   explaining why this minimal solution is sufficient _now_.
 
 ## Future-Proofing Requires Evidence + Discussion
 

.cursor/rules/development/asset_configuration.mdc

@@ -9,8 +9,8 @@ alwaysApply: false
 **Date**: 2025-08-19
 **Status**: 🎯 **ACTIVE** - Asset management guidelines
 
-*Scope: Assets Only (icons, splashes, image pipelines) — not overall build
-orchestration*
+_Scope: Assets Only (icons, splashes, image pipelines) — not overall build
+orchestration_
 
 ## Intent
 

.cursor/rules/development/time.mdc

@@ -40,7 +40,7 @@ feature development, issue investigations, ADRs, and documentation**.
 
   `2025-08-17`).
 
-- Avoid ambiguous terms like *recently*, *last month*, or *soon*.
+- Avoid ambiguous terms like _recently_, _last month_, or _soon_.
 
 - For time-based experiments (e.g., A/B tests), always include:
 

.cursor/rules/development/time_implementation.mdc

@@ -19,7 +19,7 @@
 
   - Optionally provide UTC alongside if context requires cross-team clarity.
 
-- When interpreting relative terms like *now*, *today*, *last week*:
+- When interpreting relative terms like _now_, _today_, _last week_:
 
   - Resolve them against the **developer's current time**.
 

.cursor/rules/docs/documentation_references_model_agents.mdc

@@ -12,19 +12,23 @@ To ensure clarity, efficiency, and high-value documentation within code and proj
 ---
 
 ## 1. Documentation and References Must Add Clear Value
-- Only include documentation, comments, or reference links when they provide *new, meaningful information* that assists understanding or decision-making.  
+
+- Only include documentation, comments, or reference links when they provide _new, meaningful information_ that assists understanding or decision-making.  
 - Avoid duplicating content already obvious in the codebase, version history, or linked project documents.
 
 ---
 
 ## 2. Eliminate Redundant or Noisy References
+
 - Remove references that serve no purpose beyond filling space.  
 - Model agents may automatically flag and suggest removal of trivial references (e.g., links to unchanged boilerplate or self-evident context).
 
 ---
 
 ## 3. Explicit Role of Model Agents
+
 Model agents are **active participants** in documentation quality control. Their tasks include:  
+
 - **Relevance Evaluation**: Automatically analyze references for their substantive contribution before inclusion.  
 - **Redundancy Detection**: Flag duplicate or trivial references across commits, files, or tasks.  
 - **Context Linking**: Suggest appropriate higher-level docs (designs, ADRs, meeting notes) when a code change touches multi-stage or cross-team items.  
@@ -36,6 +40,7 @@ Contributors must treat agent recommendations as **first-pass reviews** but rema
 ---
 
 ## 4. Contextual References for Complex Items
+
 - Use **centralized references** for multi-stage features (e.g., architectural docs, research threads).  
 - Keep inline code comments light; push broader context into centralized documents.  
 - Model agents may auto-summarize complex chains of discussion and attach them as a single reference point.
@@ -43,24 +48,28 @@ Contributors must treat agent recommendations as **first-pass reviews** but rema
 ---
 
 ## 5. Centralization of Broader Context
+
 - Store overarching context (design docs, proposals, workflows) in accessible, well-indexed places.  
 - Model agents should assist by **generating reference maps** that track where docs are cited across the codebase.
 
 ---
 
 ## 6. Focused Documentation
+
 - Documentation should explain **why** and **how** decisions are made, not just what was changed.  
 - Model agents can auto-generate first-pass explanations from commit metadata, diffs, and linked issues—but humans must refine them for accuracy and intent.
 
 ---
 
 ## 7. Review and Accountability
+
 - Reviewers and team leads must reject submissions containing unnecessary or low-quality documentation.  
 - Model agent outputs are aids, not replacements—contributors remain responsible for **final clarity and relevance**.
 
 ---
 
 ## 8. Continuous Improvement and Agent Feedback Loops
+
 - Encourage iterative development of model agents so their evaluations become more precise over time.  
 - Contributions should include **feedback on agent suggestions** (e.g., accepted, rejected, or corrected) to train better future outputs.  
 - Agents should log patterns of “rejected” suggestions for refinement.  
@@ -84,4 +93,4 @@ flowchart TD
 
 ---
 
-✅ **Outcome:** By integrating disciplined contributor standards with **model agent augmentation**, the team achieves documentation that is consistently *relevant, concise, centralized, and decision-focused*. AI ensures coverage and noise reduction, while humans ensure precision and judgment.  
+✅ **Outcome:** By integrating disciplined contributor standards with **model agent augmentation**, the team achieves documentation that is consistently _relevant, concise, centralized, and decision-focused_. AI ensures coverage and noise reduction, while humans ensure precision and judgment.  

.cursor/rules/harbor_pilot_universal.mdc

@@ -16,7 +16,7 @@ inherits: base_context.mdc
 
 **Author**: System/Shared  
 **Date**: 2025-08-21 (UTC)  
-**Status**: 🚢 ACTIVE — General ruleset extending *Base Context — Human Competence First*
+**Status**: 🚢 ACTIVE — General ruleset extending _Base Context — Human Competence First_
 
 > **Alignment with Base Context**
 >
@@ -40,7 +40,7 @@ Produce a **developer-grade, reproducible guide** for any technical topic that o
   - **APIs**: `curl` + one client library (e.g., `httpx` for Python).
   - **CLIs**: literal command blocks and expected output snippets.
   - **Code**: minimal, self-contained samples (language appropriate).
-- Cite **evidence** for *Works/Doesn’t* items (timestamps, filenames, line numbers, IDs/status codes, or logs).
+- Cite **evidence** for _Works/Doesn’t_ items (timestamps, filenames, line numbers, IDs/status codes, or logs).
 - If something is unknown, output `TODO:<missing>` — **never invent**.
 
 ## Required Sections (extends Base Output Contract)
@@ -56,9 +56,9 @@ Follow this exact order **after** the Base Contract’s **Objective → Result
 4. **Architecture / Process Overview**
    - Short prose + **one diagram** selected from the list above.
 5. **Interfaces & Contracts (choose one)**
-   - **API-based**: Endpoint table (*Step, Method, Path/URL, Auth, Key Headers/Params, Sample Req/Resp ref*).
-   - **Data/Files**: I/O contract table (*Source, Format, Schema/Columns, Size, Validation rules*).
-   - **Systems/Hardware**: Interfaces table (*Port/Bus, Protocol, Voltage/Timing, Constraints*).
+   - **API-based**: Endpoint table (_Step, Method, Path/URL, Auth, Key Headers/Params, Sample Req/Resp ref_).
+   - **Data/Files**: I/O contract table (_Source, Format, Schema/Columns, Size, Validation rules_).
+   - **Systems/Hardware**: Interfaces table (_Port/Bus, Protocol, Voltage/Timing, Constraints_).
 6. **Repro: End-to-End Procedure**
    - Minimal copy-paste steps with code/commands and **expected outputs**.
 7. **What Works (with Evidence)**
@@ -74,10 +74,10 @@ Follow this exact order **after** the Base Contract’s **Objective → Result
 
 > **Competence Hooks (per Base Context; keep lightweight):**
 >
-> - *Why this works* (≤3 bullets) — core invariants or guarantees.  
-> - *Common pitfalls* (≤3 bullets) — the traps we saw in evidence.  
-> - *Next skill unlock* (1 line) — the next capability to implement/learn.  
-> - *Teach-back* (1 line) — prompt the reader to restate the flow/architecture.
+> - _Why this works_ (≤3 bullets) — core invariants or guarantees.  
+> - _Common pitfalls_ (≤3 bullets) — the traps we saw in evidence.  
+> - _Next skill unlock_ (1 line) — the next capability to implement/learn.  
+> - _Teach-back_ (1 line) — prompt the reader to restate the flow/architecture.
 
 > **Collaboration Hooks (per Base Context):**
 >
@@ -203,10 +203,10 @@ Follow this exact order **after** the Base Contract’s **Objective → Result
 
 ## Competence Hooks
 
-- *Why this works*: <≤3 bullets>
-- *Common pitfalls*: <≤3 bullets>
-- *Next skill unlock*: <1 line>
-- *Teach-back*: <1 line>
+- _Why this works_: <≤3 bullets>
+- _Common pitfalls_: <≤3 bullets>
+- _Next skill unlock_: <1 line>
+- _Teach-back_: <1 line>
 
 ## Collaboration Hooks
 
@@ -226,7 +226,7 @@ Follow this exact order **after** the Base Contract’s **Objective → Result
 
 **Notes for Implementers:**  
 
-- Respect Base *Do-Not* (no filler, no invented facts, no censorship).  
+- Respect Base _Do-Not_ (no filler, no invented facts, no censorship).  
 - Prefer clarity over completeness when timeboxed; capture unknowns explicitly.
 - Apply historical comment management rules (see `.cursor/rules/historical_comment_management.mdc`)
 - Apply realistic time estimation rules (see `.cursor/rules/realistic_time_estimation.mdc`)

.cursor/rules/workflow/commit_messages.mdc

@@ -73,7 +73,7 @@
 
 ### Avoid
 
-- Vague: *improved, enhanced, better*
+- Vague: _improved, enhanced, better_
 
 - Trivialities: tiny docs, one-liners, pure lint cleanups (separate,