prr-kit 1.1.3 → 1.2.0

package/src/prr/workflows/6-report/post-comments/steps/step-02-post.md

@@ -9,41 +9,70 @@ description: "Post comments via platform CLI/API, verify, clean up"
 
 ### 1. Post Based on Platform
 
+**⚠️ Always use `{gh_path}` (the resolved full path from step-01), not bare `gh`.**
+**⚠️ Always quote paths containing spaces.**
+
+---
+
 **GitHub:**
 ```bash
-gh api repos/{platform_repo}/pulls/{pr_number}/reviews \
+"{gh_path}" api repos/{platform_repo}/pulls/{pr_number}/reviews \
   --method POST \
   --header "Accept: application/vnd.github+json" \
-  --input {review_output}/.prr-payload.json
+  --input "{temp_dir}/prr-payload.json"
+```
+
+**Handle 422 errors — two known cases:**
+
+**Case A — Self-review** (`"Review Can not request changes on your own pull request"`):
+- The authenticated user is the PR author. GitHub disallows `REQUEST_CHANGES` on own PRs.
+- Fix: Update `event` in the payload from `"REQUEST_CHANGES"` to `"COMMENT"` and retry once.
+- Use whichever runtime is available (`{detected_runtime}` from step A earlier):
+
+```bash
+# Node.js
+node -e "const fs=require('fs'); const p=JSON.parse(fs.readFileSync('{temp_dir}/prr-payload.json','utf-8')); p.event='COMMENT'; fs.writeFileSync('{temp_dir}/prr-payload.json',JSON.stringify(p,null,2)); console.log('event → COMMENT')"
+
+# Python3
+python3 -c "import json; f=open('{temp_dir}/prr-payload.json','r+'); p=json.load(f); p['event']='COMMENT'; f.seek(0); json.dump(p,f,ensure_ascii=False,indent=2); f.truncate(); print('event → COMMENT')"
 ```
+Then retry the POST above.
+
+**Case B — Line not in diff** (`"Pull Request Review thread not found"` / line outside diff hunk):
+- Remove the offending comment from the payload (the one with the invalid `path`/`line`), append its body to the summary `body` field as a fallback, then retry.
+- Repeat until all comments are either posted inline or moved to summary.
 
-Handle 422 (line not in diff): remove the offending inline comment from payload, add to summary body, retry.
+---
 
 **GitLab:**
 ```bash
 # Post summary as MR note
-glab mr note {pr_number} --repo {platform_repo} --message "$(cat {review_output}/.prr-summary.md)"
+glab mr note {pr_number} --repo {platform_repo} --message "$(cat '{temp_dir}/prr-summary.md')"
 
 # Post each inline comment
-for payload in {review_output}/.prr-payload-*.json; do
+for payload in "{temp_dir}/prr-payload-"*.json; do
   glab api projects/{encoded_repo}/merge_requests/{pr_number}/discussions \
     --method POST --input "$payload"
 done
 ```
 Where `{encoded_repo}` = `{platform_repo}` with `/` replaced by `%2F`.
 
+---
+
 **Azure DevOps:**
 ```bash
 # Post summary as PR comment
-az repos pr comment add --id {pr_number} --comment "$(cat {review_output}/.prr-summary.md)"
+az repos pr comment add --id {pr_number} --comment "$(cat '{temp_dir}/prr-summary.md')"
 
 # Post each inline thread
-for payload in {review_output}/.prr-thread-*.json; do
+for payload in "{temp_dir}/prr-thread-"*.json; do
   az repos pr thread create --id {pr_number} \
     --template-file "$payload"
 done
 ```
 
+---
+
 **Bitbucket:**
 ```bash
 # Post summary
@@ -51,10 +80,10 @@ curl -s -X POST \
   "https://api.bitbucket.org/2.0/repositories/{platform_repo}/pullrequests/{pr_number}/comments" \
   -H "Authorization: Bearer {BB_TOKEN}" \
   -H "Content-Type: application/json" \
-  -d "{\"content\": {\"raw\": \"$(cat {review_output}/.prr-summary.md)\"}}"
+  -d "{\"content\": {\"raw\": \"$(cat '{temp_dir}/prr-summary.md')\"}}"
 
 # Post each inline comment
-for payload in {review_output}/.prr-bb-*.json; do
+for payload in "{temp_dir}/prr-bb-"*.json; do
   curl -s -X POST \
     "https://api.bitbucket.org/2.0/repositories/{platform_repo}/pullrequests/{pr_number}/comments" \
     -H "Authorization: Bearer {BB_TOKEN}" \
@@ -63,25 +92,67 @@ for payload in {review_output}/.prr-bb-*.json; do
 done
 ```
 
-### 2. Verify
+---
+
+### 2. Construct PR/MR URL
+
+Build `{pr_url}` from available variables based on `{active_platform}`:
+
+| Platform | URL pattern |
+|----------|------------|
+| `github` | `https://github.com/{platform_repo}/pull/{pr_number}` |
+| `gitlab` | `https://gitlab.com/{platform_repo}/-/merge_requests/{pr_number}` |
+| `azure` | `https://dev.azure.com/{org}/{project}/_git/{repo}/pullrequest/{pr_number}` |
+| `bitbucket` | `https://bitbucket.org/{platform_repo}/pull-requests/{pr_number}` |
+| `none` | *(not applicable)* |
+
+For Azure, extract `{org}`, `{project}`, `{repo}` from `{platform_repo}` (format: `org/project/repo`).
+
+If `{pr_number}` is empty (branch-only flow), set `{pr_url}` = *(not available)*.
+
+---
+
+### 3. Verify
+
+Confirm the post was successful.
+
+**⚠️ Do NOT pipe CLI output to `node -e` or `python -e` via stdin on Windows** — `/dev/stdin` does not exist. Use `--jq` (gh CLI built-in) or write output to a temp file first.
+
+**GitHub:**
+```bash
+"{gh_path}" pr view {pr_number} --repo {platform_repo} --json reviews --jq '.reviews[-1].state'
+```
+Expected: `"COMMENTED"` or `"APPROVED"` or `"CHANGES_REQUESTED"`.
+
+**GitLab:**
+```bash
+glab mr view {pr_number} --repo {platform_repo} --output json | jq '.user_notes_count'
+```
+
+**Azure:**
+```bash
+az repos pr thread list --id {pr_number} --query "length(@)"
+```
 
-Confirm the post was successful by checking for a `200` or `201` response.
+**Bitbucket:** check response `id` field from the POST above.
 
-**GitHub:** `gh pr view {pr_number} --repo {platform_repo} --json reviews --jq '.reviews[-1].state'`
-**GitLab:** `glab mr view {pr_number} --repo {platform_repo} --output json | jq '.user_notes_count'`
-**Azure:** `az repos pr thread list --id {pr_number} --query "length(@)"`
-**Bitbucket:** check response `id` field from the post
+---
 
-### 3. Clean Up Temp Files
+### 4. Clean Up Temp Files
 
 ```bash
-rm -f {review_output}/.prr-payload*.json \
-       {review_output}/.prr-summary.md \
-       {review_output}/.prr-thread-*.json \
-       {review_output}/.prr-bb-*.json
+rm -f "{temp_dir}/prr-payload.json" \
+       "{temp_dir}/prr-payload-"*.json \
+       "{temp_dir}/prr-summary.md" \
+       "{temp_dir}/prr-thread-"*.json \
+       "{temp_dir}/prr-bb-"*.json \
+       "{temp_dir}/build-payload.mjs" \
+       "{temp_dir}/build-payload.py"
 ```
 
-### 4. Display Completion
+---
+
+### 5. Display Completion
 
 ```
 ✅ Review Posted!

package/src/prr/workflows/6-report/post-comments/workflow.md

@@ -40,6 +40,12 @@ Load config from `{main_config}`.
 Load PR context from `{review_output}/current-pr-context.yaml`.
 Use `{active_platform}` (or detect from `{platform}` config).
 
+Set `{temp_dir}` = `{review_output}/temp`.
+Create it if it doesn't exist:
+```bash
+mkdir -p "{temp_dir}"
+```
+
 ## EXECUTION
 
 Read fully and follow: `{nextStep}`

package/src/prr/workflows/quick/workflow.md

@@ -11,12 +11,12 @@ Only pause for user input when selecting the branch. Everything else runs automa
 
 ## INITIALIZATION
 
-Load config from {main_config}: `user_name`, `communication_language`, `target_repo`, `review_output`, `project_context`.
+Load config from {main_config}: `user_name`, `communication_language`, `target_repo`, `review_output`, `auto_post_comment`.
 
 Set `date` = today's date (YYYY-MM-DD).
 
-If `project_context` file exists at `{review_output}/project-context.yaml`, load it silently.
-If it does not exist, notify: "⚠️ No project context found — run [CP] Collect Project Context first for better reviews. Continuing without it."
+**Note:** Context will be collected dynamically in Phase 2.5 after describing the PR.
+No pre-collected context file needed. Context is always fresh and PR-specific.
 
 ---
 
@@ -29,13 +29,33 @@ git -C {target_repo} fetch origin --prune
 ```
 Show: `✓ Fetched latest from remote`
 
-### 1b. List open PRs (primary) + recent branches (secondary)
+### 1b. List open PRs/MRs (primary) + recent branches (secondary)
 
-**Primary — GitHub PRs** (if `{github_repo}` is configured):
+**Primary — Platform PRs/MRs** (if `{platform_repo}` is configured):
+
+Use the appropriate command based on `{platform}` (or `{detected_platform}`):
+
+**GitHub:**
 ```bash
-gh pr list --repo {github_repo} --state open \
+gh pr list --repo {platform_repo} --state open \
   --json number,title,headRefName,baseRefName,author,createdAt,isDraft --limit 20
 ```
+
+**GitLab:**
+```bash
+glab mr list --repo {platform_repo} --state opened --output json
+```
+
+**Azure DevOps:**
+```bash
+az repos pr list --repository {repo} --project {project} --org {org_url} --status active --output json
+```
+
+**Bitbucket:**
+```bash
+curl https://api.bitbucket.org/2.0/repositories/{platform_repo}/pullrequests?state=OPEN
+```
+
 Display as a table: `#N | title | head → base | author | age`
 
 **Secondary — recent branches** (always):
@@ -45,29 +65,49 @@ git -C {target_repo} branch -r --sort=-committerdate \
   | head -15
 ```
 
-### 1c. Select PR ← **ONLY USER INPUT IN THIS WORKFLOW**
+### 1c. Select PR/MR ← **ONLY USER INPUT IN THIS WORKFLOW**
 
-**If `{github_repo}` is configured** — ask:
-> Select a PR to review:
-> Enter PR number (e.g. `44`) or branch name (e.g. `feature/my-feature`):
+**If `{platform_repo}` is configured** — ask:
+> Select a PR/MR to review:
+> Enter PR/MR number (e.g. `44`) or branch name (e.g. `feature/my-feature`):
 
 Wait for response.
 
-**If PR number entered:**
+**If PR/MR number entered**, fetch metadata using the active platform:
+
+**GitHub:**
 ```bash
-gh pr view {pr_number} --repo {github_repo} \
+gh pr view {pr_number} --repo {platform_repo} \
   --json number,title,headRefName,baseRefName,author,headRefOid
 ```
-Set `target_branch` = `headRefName`, `base_branch` = `baseRefName` ← **exact from GitHub, not assumed**.
+Set `target_branch` = `headRefName`, `base_branch` = `baseRefName` ← **exact from platform, not assumed**.
 Set `pr_head_sha` = `headRefOid`.
 
+**GitLab:**
+```bash
+glab mr view {pr_number} --repo {platform_repo} --output json
+```
+Set `target_branch` = `source_branch`, `base_branch` = `target_branch`.
+Get head SHA: `git -C {target_repo} rev-parse origin/{target_branch}` → `pr_head_sha`.
+
+**Azure DevOps:**
+```bash
+az repos pr show --id {pr_number} --output json
+```
+Set `target_branch` = `sourceRefName` (strip `refs/heads/`), `base_branch` = `targetRefName` (strip `refs/heads/`).
+
+**Bitbucket:**
+```bash
+curl https://api.bitbucket.org/2.0/repositories/{platform_repo}/pullrequests/{pr_number}
+```
+Set `target_branch` = `source.branch.name`, `base_branch` = `destination.branch.name`.
+
 **If branch name entered:**
-Check if a PR exists for it via `gh pr list --head {branch}`.
-If yes: use PR's `baseRefName`. If no: detect `origin/main` or `origin/master`.
+Check if a PR/MR exists for it on the platform. If yes: use its base branch. If no: detect `origin/main` or `origin/master`.
 
 ---
 
-**If `{github_repo}` is NOT configured** — ask two separate questions:
+**If `{platform_repo}` is NOT configured** — ask two separate questions:
 
 First, display EXACTLY:
 ```
@@ -92,11 +132,21 @@ Set `base_branch` = input or detected default.
 Set `diff_range` = `{base_branch}...{target_branch}`.
 
 ### 1d. Load diff
+
+Use the first available method based on platform:
+
+**GitHub** (if `active_platform = github` and `pr_number` is set):
 ```bash
-# If PR number available (preferred):
-gh pr diff {pr_number} --repo {github_repo}
+gh pr diff {pr_number} --repo {platform_repo}
+```
 
-# Otherwise:
+**GitLab** (if `active_platform = gitlab` and `pr_number` is set):
+```bash
+glab mr diff {pr_number} --repo {platform_repo}
+```
+
+**Azure DevOps / Bitbucket / fallback:**
+```bash
 git -C {target_repo} diff {base_branch}...{target_branch} --stat
 git -C {target_repo} diff {base_branch}...{target_branch}
 ```
@@ -105,9 +155,17 @@ Store diff in memory. Count files changed, lines added/removed.
 ### 1e. Save PR context
 Write `{review_output}/current-pr-context.yaml`:
 ```yaml
-target_branch: "{target_branch}"
-base_branch: "{base_branch}"
-date: "{date}"
+pr:
+  target_branch: "{target_branch}"
+  base_branch: "{base_branch}"
+  pr_number: "{pr_number}"
+  pr_title: "{pr_title}"
+  platform: "{active_platform}"
+  platform_repo: "{platform_repo}"
+  date: "{date}"
+review:
+  completed: []
+  findings: []
 ```
 
 Show summary:
@@ -122,7 +180,7 @@ Show summary:
 *Execute automatically, no user input.*
 
 ### 2a. Classify PR type
-Analyze the diff and classify as one of: `bugfix` | `feature` | `refactor` | `docs` | `test` | `chore` | `hotfix`
+Analyze the diff and classify as one of: `bugfix` | `feature` | `refactor` | `performance` | `security` | `docs` | `test` | `config` | `chore` | `hotfix`
 
 ### 2b. Generate walkthrough
 For each changed file, write a 1-2 sentence summary of what changed and why.
@@ -143,41 +201,73 @@ Print to screen:
 
 ---
 
+## PHASE 2.5 — COLLECT PR-SPECIFIC CONTEXT
+*Execute automatically, no user input.*
+
+Execute the collect-pr-context workflow in full:
+`{project-root}/_prr/prr/workflows/2-analyze/collect-pr-context/workflow.md`
+
+This workflow analyzes changed files, detects technology stacks, collects relevant context from all sources (primary docs, config files, standards docs, inline annotations, stack-specific rules, external MCP/RAG tools), and builds a structured PR-specific knowledge base at `{review_output}/pr-{pr_number}-context.yaml`.
+
+On completion, store `pr_knowledge_base` = path to the generated context file.
+
+---
+
 ## PHASE 3 — REVIEW
 *Execute all review types automatically, one by one.*
 
-For each review, read the corresponding instructions file and apply it to `{pr_diff}`.
+**Before each review, set these variables so the review instructions resolve correctly:**
+- `pr_context` = `{review_output}/current-pr-context.yaml`
+- `target_branch` = `pr_context.pr.target_branch`
+- `base_branch` = `pr_context.pr.base_branch`
+- `pr_number` = `pr_context.pr.pr_number`
+- `pr_knowledge_base` = path stored in `pr_context` (set by Phase 2.5)
+- `target_repo`, `communication_language` = from config
+- `output_file` = per-review path defined below *(ensures findings are saved to disk for [RR] and [PC] later)*
 
 ### 3a. General Review
+Set `output_file` = `{review_output}/general-review-{date}.md`
 Load and follow: `{project-root}/_prr/prr/workflows/3-review/general-review/instructions.xml`
 
 Collect findings as `{general_findings}`.
 Print section header: `## 👁️ General Review`
 
 ### 3b. Security Review
+Set `output_file` = `{review_output}/security-review-{date}.md`
 Load and follow: `{project-root}/_prr/prr/workflows/3-review/security-review/instructions.xml`
 
 Collect findings as `{security_findings}`.
 Print section header: `## 🔒 Security Review`
 
 ### 3c. Performance Review
+Set `output_file` = `{review_output}/performance-review-{date}.md`
 Load and follow: `{project-root}/_prr/prr/workflows/3-review/performance-review/instructions.xml`
 
 Collect findings as `{performance_findings}`.
 Print section header: `## ⚡ Performance Review`
 
 ### 3d. Architecture Review
+Set `output_file` = `{review_output}/architecture-review-{date}.md`
 Load and follow: `{project-root}/_prr/prr/workflows/3-review/architecture-review/instructions.xml`
 
 Collect findings as `{architecture_findings}`.
 Print section header: `## 🏗️ Architecture Review`
 
+### 3e. Business Review
+Set `output_file` = `{review_output}/business-review-{date}.md`
+Load and follow: `{project-root}/_prr/prr/workflows/3-review/business-review/instructions.xml`
+
+Collect findings as `{business_findings}`.
+Print section header: `## 💼 Business Review`
+
+**Note:** Business Review runs last — it references and translates findings from GR/SR/PR/AR into business language and user impact.
+
 ---
 
 ## PHASE 4 — GENERATE REPORT
 *Execute automatically.*
 
-Compile all findings from phases 3a–3d.
+Compile all findings from phases 3a–3e.
 
 Sort by severity: 🔴 Blockers first → 🟡 Warnings → 🟢 Suggestions → 📌 Questions.
 
@@ -198,12 +288,22 @@ Report format:
 **Type:** {pr_type} | **Files:** X | **Lines:** +Y/-Z
 
 ## Executive Summary
-{2-3 sentence overall quality assessment}
+{2-3 sentence overall technical quality assessment}
+
+**Verdict:** ✅ APPROVE / ⚠️ APPROVE WITH NOTES / 🚫 REQUEST CHANGES
+**Business Risk:** CRITICAL / HIGH / MEDIUM / LOW / MINIMAL  *(if Business Review ran)*
 
-**Totals:** 🔴 {blocker_count} blockers | 🟡 {warning_count} warnings | 🟢 {suggestion_count} suggestions
+**Totals:** 🔴 {blocker_count} blockers | 🟡 {warning_count} warnings | 🟢 {suggestion_count} suggestions | ❓ {question_count} questions
+
+## Business Impact 💼
+*(Include this section only if Business Review [3e] was completed)*
+**Risk Level:** {risk_level}
+**Top concerns:** {top_3_business_concerns}
+**Deployment recommendation:** {ship_now | ship_with_fixes | do_not_ship}
+**Post-ship monitoring:** {what_to_watch}
 
 ## Blockers 🔴
-{all blocker findings}
+{all blocker findings — sorted by category: 🔒 Security · ⚡ Performance · 🏗️ Architecture · 👁️ General · 💼 Business}
 
 ## Warnings 🟡
 {all warning findings}
@@ -212,10 +312,10 @@ Report format:
 {all suggestion findings}
 
 ## Questions 📌
-{all questions}
+{all questions for author}
 
 ## Files Reviewed
-{file list}
+{file list with issue counts per file; highlight files with 3+ findings}
 ```
 
 ---
@@ -236,8 +336,14 @@ Report:  {review_output}/review-{target_branch_slug}-{date}.md
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 ```
 
-Then ask:
-> Post these findings as inline comments to GitHub? (requires `gh` CLI and `github_repo` configured)
+**If `auto_post_comment: true`** (from config):
+→ Skip prompt. Automatically proceed to post comments.
+→ Load and follow: `{project-root}/_prr/prr/workflows/6-report/post-comments/workflow.md`
+
+**If `auto_post_comment: false`** (default):
+→ Ask:
+> Post these findings as inline comments to the PR/MR? (requires platform CLI and `platform_repo` configured)
+> Supports: GitHub (`gh`), GitLab (`glab`), Azure DevOps (`az`), Bitbucket (API)
 > Type **PC** to post, or **Enter** to finish.
 
 If user types `PC`, load and follow: `{project-root}/_prr/prr/workflows/6-report/post-comments/workflow.md`

package/src/prr/workflows/0-setup/collect-project-context/steps/step-01-scan-configs.md

@@ -1,106 +0,0 @@
----
-name: "step-01-scan-configs"
-description: "Scan the target repo for config files, linting rules, and standards documents"
-nextStepFile: "./step-02-extract-rules.md"
----
-
-# Step 1: Scan Repository Config Files
-
-## Sequence of Instructions
-
-### 1. Announce Scan
-
-```
-🔍 Scanning project: {project_name}
-   Repo: {target_repo}
-   Looking for: config files, standards docs, architecture references...
-```
-
-### 2. Scan for Linting & Formatting Configs
-
-Check for these files in `{target_repo}` (root and common subdirs):
-
-**JavaScript/TypeScript:**
-- `.eslintrc`, `.eslintrc.js`, `.eslintrc.json`, `.eslintrc.yaml`, `eslint.config.mjs`, `eslint.config.js`
-- `.prettierrc`, `.prettierrc.json`, `prettier.config.mjs`, `prettier.config.js`
-- `tsconfig.json`, `tsconfig.*.json`
-- `.editorconfig`
-
-**Python:**
-- `pyproject.toml` (look for `[tool.ruff]`, `[tool.black]`, `[tool.isort]`, `[tool.flake8]`)
-- `setup.cfg` (look for `[flake8]`, `[mypy]`)
-- `.flake8`, `mypy.ini`
-
-**CSS/Vue/React:**
-- `.stylelintrc`, `stylelint.config.js`
-- `vite.config.js`, `vite.config.ts`, `vue.config.js`
-
-**General:**
-- `.editorconfig`
-- `sonar-project.properties`
-- `.pre-commit-config.yaml`
-
-For each found file: read it and note the key rules.
-
-### 3. Scan for Standards Documents
-
-Look for these files anywhere in `{target_repo}`:
-
-```
-CONTRIBUTING.md
-CONTRIBUTING.rst
-DEVELOPMENT.md
-CODING_STANDARDS.md
-CODING_STYLE.md
-CODE_STYLE.md
-ARCHITECTURE.md
-ARCHITECTURE_DECISION*.md
-docs/architecture/
-docs/adr/          ← Architecture Decision Records
-docs/standards/
-docs/conventions/
-.github/CONTRIBUTING.md
-```
-
-For each found file: read the relevant sections (skip boilerplate like "how to submit a PR").
-
-Extract from these docs:
-- Named conventions (e.g. "we use PascalCase for components")
-- Prohibited patterns (e.g. "never use `var`", "no direct DOM manipulation")
-- Required patterns (e.g. "all API calls go through the service layer")
-- Domain terminology that has specific meaning
-
-### 4. Scan for Package/Dependency Info
-
-Read `package.json` (or `requirements.txt` / `pyproject.toml` / `Cargo.toml` / `go.mod`):
-
-Extract:
-- Main framework and version (Vue 3 / React 18 / Express 4 / etc.)
-- Key libraries that imply patterns (e.g. `pinia` → state management pattern, `prisma` → ORM layer)
-- Test framework (jest / vitest / pytest / etc.)
-- Build toolchain (vite / webpack / esbuild / etc.)
-
-### 5. Detect Project Type
-
-Based on files found and dependencies, classify:
-
-**Frontend framework:** Vue 3 / React / Angular / Svelte / vanilla / none
-**Backend framework:** Express / Fastify / NestJS / Django / FastAPI / Spring / none
-**Language:** TypeScript / JavaScript / Python / Java / Go / Rust / mixed
-**DB/ORM:** Prisma / TypeORM / Sequelize / SQLAlchemy / none
-**Test runner:** Vitest / Jest / Pytest / JUnit / none
-**State management:** Pinia / Vuex / Redux / Zustand / none
-
-### 6. Report What Was Found
-
-```
-✅ Scan complete:
-   📄 Config files found: {n} ({list of filenames})
-   📚 Standards docs found: {n} ({list of filenames})
-   🏗️  Detected stack: {frontend} + {backend} ({language})
-   ⚠️  Not found: {list of expected but missing files}
-```
-
-### 7. Load Next Step
-
-Add `step-01-scan-configs` to `stepsCompleted`. Load: `{nextStepFile}`