reprompt-cli 1.6.0__tar.gz → 1.8.0__tar.gz

{reprompt_cli-1.6.0 → reprompt_cli-1.8.0}/.gitignore

@@ -41,6 +41,7 @@ thoughts/
 .tldr/
 .tldrignore
 .worktrees/
+CLAUDE.md
 
 # Launch / marketing drafts (not part of the codebase)
 docs/launch/

reprompt_cli-1.8.0/.pre-commit-hooks.yaml

@@ -0,0 +1,26 @@
+- id: reprompt-lint
+  name: reprompt lint
+  description: Check AI prompt quality in session files
+  entry: reprompt lint
+  language: python
+  pass_filenames: false
+  always_run: true
+  additional_dependencies: ["reprompt-cli"]
+
+- id: reprompt-lint-strict
+  name: reprompt lint (strict)
+  description: Check AI prompt quality — fail on warnings too
+  entry: reprompt lint --strict
+  language: python
+  pass_filenames: false
+  always_run: true
+  additional_dependencies: ["reprompt-cli"]
+
+- id: reprompt-lint-score
+  name: reprompt lint (score gate)
+  description: Check AI prompt quality — fail if avg score below threshold
+  entry: reprompt lint --score-threshold 50
+  language: python
+  pass_filenames: false
+  always_run: true
+  additional_dependencies: ["reprompt-cli"]

{reprompt_cli-1.6.0 → reprompt_cli-1.8.0}/CHANGELOG.md

@@ -2,6 +2,22 @@
 
 All notable changes to this project will be documented in this file.
 
+## [1.6.1] - 2026-03-28
+
+### Fixed
+- **Critical: `reprompt scan` crash** — `compute_pattern_effectiveness` failed on prompts containing SQL LIKE wildcards (`%`, `_`). Replaced `LIKE` with `INSTR()` for exact substring matching.
+- **Score calibration** — single-paragraph prompts incorrectly received 0/20 for Position (instruction was treated as "buried in the middle"). Now correctly scores instruction-at-start as optimal position.
+- **Compress grammar bugs** — fixed leading whitespace after filler deletion, dangling "that" after "the thing is" removal, orphaned commas, and remnant pleasantry fragments.
+- **Deprecated command references** — `reprompt library` and `reprompt trends` references in report footer and scan guidance now point to `reprompt template list` and `reprompt digest --trends`.
+- **Wrong adapter names** — `--source` help examples used non-existent `chatgpt-ext`; fixed to `chatgpt-export`.
+
+### Improved
+- **Compress engine** — added 15 filler phrases (`kind of`, `sort of`, `additionally`, `the fact that`, etc.) and 16 phrase simplification rules (`take a look at` → `check`, `let me know` → removed, etc.). Added post-compression cleanup pass for whitespace, punctuation, and sentence capitalization. Typical savings improved from 3-33% to 38-60%.
+- **Score UX** — suggestions now sorted by impact (high first); paper citations dimmed; new "Very Poor" grade tier for scores under 20.
+- **CLI help** — added usage examples to 7 key commands (score, scan, compress, compare, distill, lint, privacy). Added Quick Start guide to main `--help`. Standardized `--source` option across all commands with `-s` shorthand.
+- **Help panels** — moved `digest` from Manage to Analyze panel. Listed all 9 adapters in `scan --help`.
+- Tests: 1490 → 1497
+
 ## [1.6.0] - 2026-03-28
 
 ### Added

reprompt_cli-1.8.0/CONTRIBUTING.md

@@ -0,0 +1,120 @@
+# Contributing to reprompt
+
+Thanks for your interest in contributing! reprompt is an open-source prompt intelligence tool, and contributions of all kinds are welcome.
+
+## Quick Links
+
+- [Good First Issues](https://github.com/reprompt-dev/reprompt/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22)
+- [Help Wanted](https://github.com/reprompt-dev/reprompt/issues?q=is%3Aissue+is%3Aopen+label%3A%22help+wanted%22)
+
+## Development Setup
+
+```bash
+git clone https://github.com/reprompt-dev/reprompt
+cd reprompt
+uv venv
+uv pip install -e ".[dev]"
+```
+
+## Running Tests
+
+```bash
+uv run pytest tests/ -v
+uv run pytest tests/ -v --cov=reprompt  # with coverage
+```
+
+## Code Style
+
+We use [ruff](https://docs.astral.sh/ruff/) for linting and formatting:
+
+```bash
+uv run ruff check src/ tests/
+uv run ruff format src/ tests/
+```
+
+## Type Checking
+
+We use [mypy](https://mypy-lang.org/) in strict mode:
+
+```bash
+uv run mypy src/reprompt/
+```
+
+## Pull Requests
+
+1. Fork the repo and create a feature branch
+2. Write tests for new functionality
+3. Ensure all tests pass and coverage doesn't decrease
+4. Run `ruff check` and `ruff format`
+5. Run `mypy src/reprompt/` -- must pass clean
+6. Submit a PR with a clear description
+
+### Before Submitting
+
+- **Comment on the issue first** — introduce yourself and outline your approach before writing code. This avoids wasted work and duplicate effort.
+- **One PR per issue** — don't submit multiple PRs for the same issue.
+- PRs from brand-new GitHub accounts with no prior project engagement will generally not be reviewed.
+
+## Common Contribution Areas
+
+### Adding Adapters (new AI tool support)
+
+Adding a new adapter is the easiest way to contribute. Each adapter is ~50-100 lines:
+
+1. Create `src/reprompt/adapters/your_tool.py`
+2. Subclass `BaseAdapter` from `adapters/base.py`
+3. Implement:
+   - `name` property -- adapter identifier (e.g., `"copilot"`)
+   - `default_session_path` -- where the tool stores sessions
+   - `discover_sessions()` -- find session files
+   - `parse_session(path)` -- extract `Prompt` objects
+   - Optional: `parse_conversation(path)` -- extract full conversation turns
+4. Register in `core/pipeline.py` → `get_adapters()`
+5. Add tests in `tests/test_your_tool_adapter.py`
+
+**Reference adapters:**
+- `adapters/cline.py` -- simplest (~80 lines, JSON format)
+- `adapters/aider.py` -- markdown parsing
+- `adapters/cursor.py` -- SQLite/vscdb format
+
+### Adding Lint Rules
+
+Each lint rule is a standalone function in `core/lint.py`:
+
+1. Add your rule check in `lint_prompt()`
+2. Return a `LintViolation` with `severity="error"` or `"warning"`
+3. Add tests in `tests/test_lint.py`
+
+### Adding Privacy Patterns
+
+Extend sensitive content detection in `core/privacy_scan.py`:
+
+1. Add a regex pattern to the `PATTERNS` dict
+2. Add category mapping in `CATEGORY_MAP`
+3. Add safety filter if needed (to exclude false positives)
+4. Add tests in `tests/test_privacy_scan.py`
+
+## Architecture
+
+```
+src/reprompt/
+├── cli.py              # Typer CLI (21 commands)
+├── config.py           # pydantic-settings configuration
+├── core/               # Business logic
+│   ├── models.py       # Prompt dataclass
+│   ├── scorer.py       # Research-calibrated 0-100 scoring
+│   ├── lint.py         # Quality linter
+│   ├── compress.py     # 4-layer prompt compression
+│   ├── distill.py      # Conversation distillation
+│   ├── agent.py        # Agent workflow analysis
+│   ├── privacy_scan.py # Sensitive content detection
+│   └── pipeline.py     # Orchestrator
+├── adapters/           # 9 AI tool parsers
+├── storage/            # SQLite layer
+├── bridge/             # Browser extension Native Messaging
+└── output/             # Report formatters (terminal, JSON, HTML)
+```
+
+## Questions?
+
+Open an issue or start a discussion. We're happy to help you get started.

{reprompt_cli-1.6.0 → reprompt_cli-1.8.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: reprompt-cli
-Version: 1.6.0
+Version: 1.8.0
 Summary: Discover, analyze, and optimize your prompts from AI coding sessions
 Project-URL: Homepage, https://github.com/reprompt-dev/reprompt
 Project-URL: Repository, https://github.com/reprompt-dev/reprompt
@@ -45,7 +45,7 @@ Description-Content-Type: text/markdown
 
 # `re:prompt`
 
-**Prompt Intelligence for AI power users** -- understand, optimize, and manage your AI conversations.
+**Analyze what you type into AI tools** -- prompt scoring, agent error loops, leaked credential detection, conversation distillation.
 
 [![PyPI version](https://img.shields.io/pypi/v/reprompt-cli)](https://pypi.org/project/reprompt-cli/)
 [![Python 3.10+](https://img.shields.io/pypi/pyversions/reprompt-cli)](https://pypi.org/project/reprompt-cli/)
@@ -55,6 +55,8 @@ Description-Content-Type: text/markdown
 
 ---
 
+![reprompt demo](docs/demo.gif)
+
 ## See it in action
 
 ```bash
@@ -67,18 +69,18 @@ $ reprompt
   ╰────────────────────────────────────────────────────────────╯
 
 $ reprompt score "Fix the auth bug in src/login.ts where JWT expires"
-  Score: 74/100
-  Structure: 18/25 | Context: 22/25 | Position: 15/20 | Repetition: 9/15 | Clarity: 10/15
-  Tip: Add the error message for +15% accuracy
+  Score: 40/100  (Fair)
+  Structure: 0/25 | Context: 8/25 | Position: 20/20 | Repetition: 0/15 | Clarity: 12/15
+  Tip: Include the error message -- debug prompts with errors are 3.7x more effective
 
 $ reprompt distill --last 3 --summary
   Session: feature-dev (42 turns, 18 important)
   Key moments: initial spec → auth module → test failures → JWT fix → passing
   Context: "Building auth system with JWT refresh tokens for Express API"
 
-$ reprompt compress "请帮我看一下这个代码，就是那个 login 的那个文件，好像有点问题"
-  Before: 31 tokens → After: 15 tokens (52% saved)
-  "看一下 login 文件的问题"
+$ reprompt compress "I was wondering if you could please help me refactor this code. Basically what I need is to split this function into smaller helpers and add error handling."
+  Before: 28 tokens → After: 14 tokens (50% saved)
+  "Help me refactor this code. Split this function into smaller helpers and add error handling."
 ```
 
 ## What it does
@@ -99,7 +101,7 @@ $ reprompt compress "请帮我看一下这个代码，就是那个 login 的那
 
 | Command | Description |
 |---------|-------------|
-| `reprompt compress "prompt"` | 4-layer prompt compression (50%+ token savings typical) |
+| `reprompt compress "prompt"` | 4-layer prompt compression (40-60% token savings typical) |
 | `reprompt distill` | Extract important turns from conversations with 6-signal scoring |
 | `reprompt distill --export` | Recover context when a session runs out -- paste into new session |
 | `reprompt lint` | Prompt quality linter with GitHub Action support |
@@ -183,12 +185,44 @@ reprompt install-hook               # adds post-session hook to Claude Code
 
 Capture prompts from ChatGPT, Claude.ai, and Gemini directly in your browser:
 
-1. **Install the extension** from [Chrome Web Store](https://chromewebstore.google.com/detail/reprompt/ojdccpagaanchmkninlbgbgemdcjckhn)
+1. **Install the extension** from [Chrome Web Store](https://chromewebstore.google.com/detail/reprompt/ojdccpagaanchmkninlbgbgemdcjckhn) or [Firefox Add-ons](https://addons.mozilla.org/addon/reprompt-cli/)
 2. **Connect to the CLI:** `reprompt install-extension`
 3. **Verify:** `reprompt extension-status`
 
 Captured prompts sync locally via Native Messaging -- nothing leaves your machine.
 
+### CI integration
+
+#### GitHub Action
+
+```yaml
+# .github/workflows/prompt-lint.yml
+- uses: reprompt-dev/reprompt@main
+  with:
+    score-threshold: 50   # fail if avg prompt score < 50
+    strict: true          # fail on warnings too
+    comment-on-pr: true   # post quality report as PR comment
+```
+
+#### pre-commit
+
+```yaml
+# .pre-commit-config.yaml
+repos:
+  - repo: https://github.com/reprompt-dev/reprompt
+    rev: v1.7.0
+    hooks:
+      - id: reprompt-lint
+```
+
+#### Direct CLI
+
+```bash
+reprompt lint --score-threshold 50  # exit 1 if avg score < 50
+reprompt lint --strict              # exit 1 on warnings
+reprompt lint --json                # machine-readable output
+```
+
 ## Privacy
 
 - All analysis runs locally. No prompts leave your machine.
@@ -202,6 +236,7 @@ Captured prompts sync locally via Native Messaging -- nothing leaves your machin
 
 - **Website:** [getreprompt.dev](https://getreprompt.dev)
 - **Chrome Extension:** [Chrome Web Store](https://chromewebstore.google.com/detail/reprompt/ojdccpagaanchmkninlbgbgemdcjckhn)
+- **Firefox Add-on:** [Firefox Add-ons](https://addons.mozilla.org/addon/reprompt-cli/)
 - **PyPI:** [reprompt-cli](https://pypi.org/project/reprompt-cli/)
 - **Changelog:** [CHANGELOG.md](CHANGELOG.md)
 - **Privacy:** [getreprompt.dev/privacy](https://getreprompt.dev/privacy)

{reprompt_cli-1.6.0 → reprompt_cli-1.8.0}/README.md

@@ -1,6 +1,6 @@
 # `re:prompt`
 
-**Prompt Intelligence for AI power users** -- understand, optimize, and manage your AI conversations.
+**Analyze what you type into AI tools** -- prompt scoring, agent error loops, leaked credential detection, conversation distillation.
 
 [![PyPI version](https://img.shields.io/pypi/v/reprompt-cli)](https://pypi.org/project/reprompt-cli/)
 [![Python 3.10+](https://img.shields.io/pypi/pyversions/reprompt-cli)](https://pypi.org/project/reprompt-cli/)
@@ -10,6 +10,8 @@
 
 ---
 
+![reprompt demo](docs/demo.gif)
+
 ## See it in action
 
 ```bash
@@ -22,18 +24,18 @@ $ reprompt
   ╰────────────────────────────────────────────────────────────╯
 
 $ reprompt score "Fix the auth bug in src/login.ts where JWT expires"
-  Score: 74/100
-  Structure: 18/25 | Context: 22/25 | Position: 15/20 | Repetition: 9/15 | Clarity: 10/15
-  Tip: Add the error message for +15% accuracy
+  Score: 40/100  (Fair)
+  Structure: 0/25 | Context: 8/25 | Position: 20/20 | Repetition: 0/15 | Clarity: 12/15
+  Tip: Include the error message -- debug prompts with errors are 3.7x more effective
 
 $ reprompt distill --last 3 --summary
   Session: feature-dev (42 turns, 18 important)
   Key moments: initial spec → auth module → test failures → JWT fix → passing
   Context: "Building auth system with JWT refresh tokens for Express API"
 
-$ reprompt compress "请帮我看一下这个代码，就是那个 login 的那个文件，好像有点问题"
-  Before: 31 tokens → After: 15 tokens (52% saved)
-  "看一下 login 文件的问题"
+$ reprompt compress "I was wondering if you could please help me refactor this code. Basically what I need is to split this function into smaller helpers and add error handling."
+  Before: 28 tokens → After: 14 tokens (50% saved)
+  "Help me refactor this code. Split this function into smaller helpers and add error handling."
 ```
 
 ## What it does
@@ -54,7 +56,7 @@ $ reprompt compress "请帮我看一下这个代码，就是那个 login 的那
 
 | Command | Description |
 |---------|-------------|
-| `reprompt compress "prompt"` | 4-layer prompt compression (50%+ token savings typical) |
+| `reprompt compress "prompt"` | 4-layer prompt compression (40-60% token savings typical) |
 | `reprompt distill` | Extract important turns from conversations with 6-signal scoring |
 | `reprompt distill --export` | Recover context when a session runs out -- paste into new session |
 | `reprompt lint` | Prompt quality linter with GitHub Action support |
@@ -138,12 +140,44 @@ reprompt install-hook               # adds post-session hook to Claude Code
 
 Capture prompts from ChatGPT, Claude.ai, and Gemini directly in your browser:
 
-1. **Install the extension** from [Chrome Web Store](https://chromewebstore.google.com/detail/reprompt/ojdccpagaanchmkninlbgbgemdcjckhn)
+1. **Install the extension** from [Chrome Web Store](https://chromewebstore.google.com/detail/reprompt/ojdccpagaanchmkninlbgbgemdcjckhn) or [Firefox Add-ons](https://addons.mozilla.org/addon/reprompt-cli/)
 2. **Connect to the CLI:** `reprompt install-extension`
 3. **Verify:** `reprompt extension-status`
 
 Captured prompts sync locally via Native Messaging -- nothing leaves your machine.
 
+### CI integration
+
+#### GitHub Action
+
+```yaml
+# .github/workflows/prompt-lint.yml
+- uses: reprompt-dev/reprompt@main
+  with:
+    score-threshold: 50   # fail if avg prompt score < 50
+    strict: true          # fail on warnings too
+    comment-on-pr: true   # post quality report as PR comment
+```
+
+#### pre-commit
+
+```yaml
+# .pre-commit-config.yaml
+repos:
+  - repo: https://github.com/reprompt-dev/reprompt
+    rev: v1.7.0
+    hooks:
+      - id: reprompt-lint
+```
+
+#### Direct CLI
+
+```bash
+reprompt lint --score-threshold 50  # exit 1 if avg score < 50
+reprompt lint --strict              # exit 1 on warnings
+reprompt lint --json                # machine-readable output
+```
+
 ## Privacy
 
 - All analysis runs locally. No prompts leave your machine.
@@ -157,6 +191,7 @@ Captured prompts sync locally via Native Messaging -- nothing leaves your machin
 
 - **Website:** [getreprompt.dev](https://getreprompt.dev)
 - **Chrome Extension:** [Chrome Web Store](https://chromewebstore.google.com/detail/reprompt/ojdccpagaanchmkninlbgbgemdcjckhn)
+- **Firefox Add-on:** [Firefox Add-ons](https://addons.mozilla.org/addon/reprompt-cli/)
 - **PyPI:** [reprompt-cli](https://pypi.org/project/reprompt-cli/)
 - **Changelog:** [CHANGELOG.md](CHANGELOG.md)
 - **Privacy:** [getreprompt.dev/privacy](https://getreprompt.dev/privacy)

reprompt_cli-1.8.0/action.yml

@@ -0,0 +1,211 @@
+name: "reprompt lint"
+description: "Check AI prompt quality in your codebase"
+branding:
+  icon: "zap"
+  color: "blue"
+
+inputs:
+  version:
+    description: "reprompt-cli version to install"
+    required: false
+    default: "latest"
+  source:
+    description: "Adapter to scan (claude-code, aider, gemini, cline, cursor, openclaw)"
+    required: false
+    default: ""
+  path:
+    description: "Path to scan for session files"
+    required: false
+    default: ""
+  strict:
+    description: "Fail on warnings too (not just errors)"
+    required: false
+    default: "false"
+  score-threshold:
+    description: "Fail if average prompt score is below this threshold (0-100)"
+    required: false
+    default: "0"
+  json:
+    description: "Output results as JSON"
+    required: false
+    default: "false"
+  comment-on-pr:
+    description: "Post quality report as PR comment (requires pull_request event)"
+    required: false
+    default: "false"
+
+outputs:
+  errors:
+    description: "Number of lint errors found"
+    value: ${{ steps.lint.outputs.errors }}
+  warnings:
+    description: "Number of lint warnings found"
+    value: ${{ steps.lint.outputs.warnings }}
+  avg-score:
+    description: "Average prompt score (when score-threshold > 0)"
+    value: ${{ steps.lint.outputs.avg_score }}
+
+runs:
+  using: "composite"
+  steps:
+    - name: Set up uv
+      uses: astral-sh/setup-uv@v7
+
+    - name: Install reprompt
+      shell: bash
+      env:
+        INPUT_VERSION: ${{ inputs.version }}
+      run: |
+        if [ "$INPUT_VERSION" = "latest" ]; then
+          uv tool install reprompt-cli
+        else
+          uv tool install "reprompt-cli==$INPUT_VERSION"
+        fi
+
+    - name: Scan sessions
+      shell: bash
+      env:
+        INPUT_SOURCE: ${{ inputs.source }}
+        INPUT_PATH: ${{ inputs.path }}
+      run: |
+        SCAN_ARGS=""
+        if [ -n "$INPUT_SOURCE" ]; then
+          SCAN_ARGS="$SCAN_ARGS --source $INPUT_SOURCE"
+        fi
+        if [ -n "$INPUT_PATH" ]; then
+          SCAN_ARGS="$SCAN_ARGS --path $INPUT_PATH"
+        fi
+        reprompt scan $SCAN_ARGS || true
+
+    - name: Run lint
+      id: lint
+      shell: bash
+      env:
+        INPUT_SOURCE: ${{ inputs.source }}
+        INPUT_PATH: ${{ inputs.path }}
+        INPUT_STRICT: ${{ inputs.strict }}
+        INPUT_SCORE_THRESHOLD: ${{ inputs.score-threshold }}
+        INPUT_JSON: ${{ inputs.json }}
+      run: |
+        LINT_ARGS=""
+        if [ -n "$INPUT_SOURCE" ]; then
+          LINT_ARGS="$LINT_ARGS --source $INPUT_SOURCE"
+        fi
+        if [ -n "$INPUT_PATH" ]; then
+          LINT_ARGS="$LINT_ARGS --path $INPUT_PATH"
+        fi
+        if [ "$INPUT_STRICT" = "true" ]; then
+          LINT_ARGS="$LINT_ARGS --strict"
+        fi
+        if [ "$INPUT_SCORE_THRESHOLD" != "0" ]; then
+          LINT_ARGS="$LINT_ARGS --score-threshold $INPUT_SCORE_THRESHOLD"
+        fi
+
+        # Always get JSON for outputs + PR comment
+        JSON_RESULT=$(reprompt lint --json $LINT_ARGS 2>/dev/null || echo '{"errors":0,"warnings":0}')
+        echo "$JSON_RESULT" > /tmp/reprompt-lint.json
+
+        echo "errors=$(echo "$JSON_RESULT" | python3 -c 'import sys,json; print(json.load(sys.stdin).get("errors",0))')" >> "$GITHUB_OUTPUT"
+        echo "warnings=$(echo "$JSON_RESULT" | python3 -c 'import sys,json; print(json.load(sys.stdin).get("warnings",0))')" >> "$GITHUB_OUTPUT"
+        echo "avg_score=$(echo "$JSON_RESULT" | python3 -c 'import sys,json; d=json.load(sys.stdin); print(d.get("score",{}).get("avg_score",0))')" >> "$GITHUB_OUTPUT"
+
+        # Run lint for display output
+        if [ "$INPUT_JSON" = "true" ]; then
+          reprompt lint --json $LINT_ARGS
+        else
+          reprompt lint $LINT_ARGS
+        fi
+
+    - name: Comment on PR
+      if: inputs.comment-on-pr == 'true' && github.event_name == 'pull_request'
+      shell: bash
+      env:
+        GH_TOKEN: ${{ github.token }}
+        PR_NUMBER: ${{ github.event.pull_request.number }}
+        REPO: ${{ github.repository }}
+      run: |
+        # Build markdown comment from lint JSON
+        COMMENT=$(python3 << 'PYEOF'
+        import json, os
+
+        with open("/tmp/reprompt-lint.json") as f:
+            data = json.load(f)
+
+        errors = data.get("errors", 0)
+        warnings = data.get("warnings", 0)
+        total = data.get("total_prompts", 0)
+        score = data.get("score", {})
+        violations = data.get("violations", [])
+
+        # Status badge
+        if errors > 0:
+            status = "🔴 **Failed**"
+        elif warnings > 0:
+            status = "🟡 **Passed with warnings**"
+        else:
+            status = "🟢 **Passed**"
+
+        lines = []
+        lines.append("<!-- reprompt-lint -->")
+        lines.append("## reprompt lint " + status)
+        lines.append("")
+        lines.append(f"| Metric | Value |")
+        lines.append(f"|--------|-------|")
+        lines.append(f"| Prompts checked | {total} |")
+        lines.append(f"| Errors | {errors} |")
+        lines.append(f"| Warnings | {warnings} |")
+
+        if score:
+            avg = score.get("avg_score", 0)
+            threshold = score.get("threshold", 0)
+            passed = "✅" if score.get("pass") else "❌"
+            lines.append(f"| Avg Score | {avg}/100 {passed} (threshold: {threshold}) |")
+            lines.append(f"| Score Range | {score.get('min_score', 0)} — {score.get('max_score', 0)} |")
+
+        # Violations (collapsible)
+        if violations:
+            lines.append("")
+            lines.append(f"<details><summary>📋 {len(violations)} violation(s)</summary>")
+            lines.append("")
+            lines.append("| Severity | Rule | Message |")
+            lines.append("|----------|------|---------|")
+            for v in violations[:20]:
+                sev = "🔴" if v["severity"] == "error" else "🟡"
+                lines.append(f"| {sev} | `{v['rule']}` | {v['message']} |")
+            if len(violations) > 20:
+                lines.append(f"| | | ... and {len(violations) - 20} more |")
+            lines.append("")
+            lines.append("</details>")
+
+        lines.append("")
+        lines.append("*Run `reprompt lint` locally for full details — [docs](https://github.com/reprompt-dev/reprompt)*")
+
+        print("\n".join(lines))
+        PYEOF
+        )
+
+        # Find existing comment to update (avoid duplicates)
+        EXISTING=$(gh api "repos/$REPO/issues/$PR_NUMBER/comments" \
+          --jq '.[] | select(.body | contains("<!-- reprompt-lint -->")) | .id' \
+          2>/dev/null | head -1)
+
+        if [ -n "$EXISTING" ]; then
+          # Update existing comment
+          gh api "repos/$REPO/issues/$PR_NUMBER/comments/$EXISTING" \
+            -X PATCH -f body="$COMMENT" > /dev/null
+        else
+          # Create new comment
+          gh api "repos/$REPO/issues/$PR_NUMBER/comments" \
+            -f body="$COMMENT" > /dev/null
+        fi
+
+    - name: Add summary
+      if: always()
+      shell: bash
+      run: |
+        echo "## reprompt lint results" >> "$GITHUB_STEP_SUMMARY"
+        echo "" >> "$GITHUB_STEP_SUMMARY"
+        echo "- Errors: ${{ steps.lint.outputs.errors }}" >> "$GITHUB_STEP_SUMMARY"
+        echo "- Warnings: ${{ steps.lint.outputs.warnings }}" >> "$GITHUB_STEP_SUMMARY"
+        echo "" >> "$GITHUB_STEP_SUMMARY"
+        echo "Run \`reprompt lint\` locally for full details." >> "$GITHUB_STEP_SUMMARY"

reprompt_cli-1.8.0/docs/icons/brand-icon.svg

@@ -0,0 +1,22 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" width="512" height="512">
+  <!--
+    reprompt Brand Icon — Production
+    Full-bleed square: let each platform apply its own mask.
+    Safe zone: content within central 75% (384x384, 64px margins).
+    Optical center: rP baseline at y=330 places visual mass at canvas center.
+  -->
+  <defs>
+    <linearGradient id="bg" x1="0" y1="0" x2="1" y2="1">
+      <stop offset="0%" stop-color="#e94560"/>
+      <stop offset="100%" stop-color="#c73050"/>
+    </linearGradient>
+  </defs>
+  <!-- Full-bleed background — NO rounded corners -->
+  <rect width="512" height="512" fill="url(#bg)"/>
+  <!-- Optically centered rP monogram -->
+  <text x="256" y="330" text-anchor="middle"
+        font-family="Inter, Helvetica Neue, Arial, -apple-system, sans-serif"
+        font-size="240">
+    <tspan fill="rgba(255,255,255,0.45)" font-weight="400">r</tspan><tspan fill="#ffffff" font-weight="800">P</tspan>
+  </text>
+</svg>