@aion0/forge 0.5.22 → 0.5.24

package/scripts/bench/tasks/01-text-utils/validator.sh

@@ -0,0 +1,46 @@
+#!/usr/bin/env bash
+# Validator for text utility task.
+# Runs in harness_test project root. Exits 0 = pass, non-zero = fail.
+set -e
+
+PROJECT_ROOT="${1:-/Users/zliu/IdeaProjects/harness_test}"
+cd "$PROJECT_ROOT/src" || { echo "FAIL: src/ directory not found"; exit 1; }
+
+# 1. Check files exist
+[ -f utils/text.js ] || { echo "FAIL: utils/text.js missing"; exit 1; }
+[ -f utils/text.test.js ] || { echo "FAIL: utils/text.test.js missing"; exit 1; }
+
+# 2. Check exports
+grep -q "export.*capitalize" utils/text.js || { echo "FAIL: capitalize not exported"; exit 1; }
+grep -q "export.*reverseWords" utils/text.js || { echo "FAIL: reverseWords not exported"; exit 1; }
+
+# 3. Run tests
+node --test utils/text.test.js 2>&1 | tee /tmp/text-test-output.txt
+TEST_EXIT=${PIPESTATUS[0]}
+if [ "$TEST_EXIT" != "0" ]; then
+  echo "FAIL: tests failed (exit=$TEST_EXIT)"
+  exit 1
+fi
+
+# 4. Additional smoke test — behavior verification independent of agent's tests
+node -e "
+import('./utils/text.js').then(m => {
+  const assert = require('node:assert/strict');
+  // capitalize
+  assert.equal(m.capitalize('hello'), 'Hello', 'capitalize basic');
+  assert.equal(m.capitalize('a'), 'A', 'capitalize single char');
+  try { m.capitalize(''); assert.fail('expected throw on empty'); } catch (e) { assert.ok(e instanceof TypeError); }
+  try { m.capitalize(null); assert.fail('expected throw on null'); } catch (e) { assert.ok(e instanceof TypeError); }
+  try { m.capitalize(123); assert.fail('expected throw on number'); } catch (e) { assert.ok(e instanceof TypeError); }
+  // reverseWords
+  assert.equal(m.reverseWords('hello world'), 'world hello');
+  assert.equal(m.reverseWords('  a  b  c  '), 'c b a');
+  assert.equal(m.reverseWords(''), '');
+  assert.equal(m.reverseWords('single'), 'single');
+  try { m.reverseWords(null); assert.fail('expected throw'); } catch (e) { assert.ok(e instanceof TypeError); }
+  console.log('SMOKE_TEST_PASSED');
+}).catch(err => { console.error('SMOKE_TEST_FAILED:', err.message); process.exit(1); });
+" || { echo "FAIL: smoke test failed"; exit 1; }
+
+echo "PASS"
+exit 0

package/scripts/bench/tasks/02-pagination/setup.sh

@@ -0,0 +1,19 @@
+#!/usr/bin/env bash
+# Create a basic user list module without pagination.
+set -e
+PROJECT="${1:-/Users/zliu/IdeaProjects/harness_test}"
+mkdir -p "$PROJECT/src/api"
+
+cat > "$PROJECT/src/api/users.js" <<'EOF'
+const USERS = Array.from({ length: 127 }, (_, i) => ({
+  id: i + 1,
+  name: `User ${i + 1}`,
+  email: `user${i + 1}@example.com`,
+}));
+
+export function listUsers() {
+  return USERS;
+}
+EOF
+
+echo "Setup complete: created src/api/users.js with 127 users and a listUsers() function."

package/scripts/bench/tasks/02-pagination/task.md

@@ -0,0 +1,48 @@
+# Task: Add Pagination to User List
+
+The file `src/api/users.js` currently has a `listUsers()` function that returns all users. Add pagination support.
+
+## Requirements
+
+Replace `listUsers()` (or add a new function) with a paginated version:
+
+```
+listUsers({ page = 1, pageSize = 20 } = {})
+```
+
+**Return format**:
+```js
+{
+  items: [...],       // users on the current page
+  total: 127,         // total number of users
+  page: 1,            // current page (1-indexed)
+  pageSize: 20,       // page size (after validation)
+  totalPages: 7,      // Math.ceil(total / pageSize)
+  hasNext: true,      // true if more pages exist
+  hasPrev: false      // true if page > 1
+}
+```
+
+## Validation Rules
+
+- `page` must be integer ≥ 1. If invalid (not a number, < 1, NaN, float), throw `RangeError`.
+- `pageSize` must be integer in [1, 100]. If invalid, throw `RangeError`.
+- If `page` exceeds available pages, return empty `items` array but still return correct `total`, `page`, `pageSize`, `totalPages`, `hasNext: false`, `hasPrev: true`.
+
+## Test File
+
+Also create `src/api/users.test.js` using `node:test` and `node:assert/strict` covering:
+- Default params return page 1 with 20 items
+- Page 2 returns items 21-40
+- Last page (page 7) returns items 121-127
+- Page 8 returns empty items but correct metadata
+- Custom pageSize (e.g., 50)
+- Invalid page (0, -1, 'abc', 1.5, NaN) throws RangeError
+- Invalid pageSize (0, 101, 'abc', 1.5) throws RangeError
+
+## Constraints
+
+- Keep ES module syntax
+- No external deps
+- Preserve the existing USERS array
+- Tests must pass via: `cd src && node --test api/users.test.js`

package/scripts/bench/tasks/02-pagination/validator.sh

@@ -0,0 +1,69 @@
+#!/usr/bin/env bash
+set -e
+PROJECT="${1:-/Users/zliu/IdeaProjects/harness_test}"
+cd "$PROJECT/src"
+
+[ -f api/users.js ] || { echo "FAIL: api/users.js missing"; exit 1; }
+[ -f api/users.test.js ] || { echo "FAIL: api/users.test.js missing"; exit 1; }
+grep -q "export function listUsers\|export const listUsers\|export { listUsers" api/users.js || { echo "FAIL: listUsers not exported"; exit 1; }
+
+# Run agent's tests
+node --test api/users.test.js 2>&1 | tee /tmp/paginate-test-output.txt
+TEST_EXIT=${PIPESTATUS[0]}
+[ "$TEST_EXIT" = "0" ] || { echo "FAIL: agent tests failed"; exit 1; }
+
+# Independent smoke test
+node -e "
+import('./api/users.js').then(m => {
+  const assert = require('node:assert/strict');
+  // Default: page 1, 20 items
+  let r = m.listUsers();
+  assert.equal(r.items.length, 20, 'default pageSize 20');
+  assert.equal(r.items[0].id, 1);
+  assert.equal(r.total, 127);
+  assert.equal(r.page, 1);
+  assert.equal(r.pageSize, 20);
+  assert.equal(r.totalPages, 7);
+  assert.equal(r.hasNext, true);
+  assert.equal(r.hasPrev, false);
+
+  // Page 2
+  r = m.listUsers({ page: 2 });
+  assert.equal(r.items[0].id, 21);
+  assert.equal(r.items.length, 20);
+  assert.equal(r.hasPrev, true);
+
+  // Last page (127 / 20 = 6.35 → 7 pages; page 7 has 7 items)
+  r = m.listUsers({ page: 7 });
+  assert.equal(r.items.length, 7, 'last page has 7 items');
+  assert.equal(r.items[0].id, 121);
+  assert.equal(r.hasNext, false);
+
+  // Page 8 (beyond) — empty but correct metadata
+  r = m.listUsers({ page: 8 });
+  assert.equal(r.items.length, 0, 'page beyond: empty items');
+  assert.equal(r.total, 127);
+  assert.equal(r.totalPages, 7);
+  assert.equal(r.hasNext, false);
+
+  // Custom pageSize
+  r = m.listUsers({ page: 1, pageSize: 50 });
+  assert.equal(r.items.length, 50);
+  assert.equal(r.totalPages, 3);
+
+  // Invalid page
+  for (const p of [0, -1, 'abc', 1.5, NaN]) {
+    try { m.listUsers({ page: p }); assert.fail('expected RangeError for page=' + p); }
+    catch (e) { assert.ok(e instanceof RangeError, 'page=' + p + ' should throw RangeError, got: ' + e.constructor.name); }
+  }
+  // Invalid pageSize
+  for (const ps of [0, 101, 'abc', 1.5]) {
+    try { m.listUsers({ page: 1, pageSize: ps }); assert.fail('expected RangeError for pageSize=' + ps); }
+    catch (e) { assert.ok(e instanceof RangeError, 'pageSize=' + ps + ' should throw RangeError'); }
+  }
+  console.log('SMOKE_TEST_PASSED');
+}).catch(err => { console.error('SMOKE_TEST_FAILED:', err.message); process.exit(1); });
+" || { echo "FAIL: smoke test failed"; exit 1; }
+
+echo "PASS"
+exit 0

package/scripts/bench/tasks/03-bug-fix/setup.sh

@@ -0,0 +1,82 @@
+#!/usr/bin/env bash
+# Create a date range calculator with 2 bugs.
+set -e
+PROJECT="${1:-/Users/zliu/IdeaProjects/harness_test}"
+mkdir -p "$PROJECT/src/lib" "$PROJECT/src/lib/__tests__"
+
+cat > "$PROJECT/src/lib/dateRange.js" <<'EOF'
+// Compute the inclusive number of days between two YYYY-MM-DD dates.
+// Returns a positive integer. If end is before start, throws RangeError.
+export function daysBetween(startStr, endStr) {
+  if (typeof startStr !== 'string' || typeof endStr !== 'string') {
+    throw new TypeError('daysBetween expects two YYYY-MM-DD strings');
+  }
+  const start = new Date(startStr);
+  const end = new Date(endStr);
+  if (Number.isNaN(start.getTime()) || Number.isNaN(end.getTime())) {
+    throw new TypeError('invalid date format');
+  }
+  if (end < start) throw new RangeError('end before start');
+  // BUG: missing +1 to be inclusive of both endpoints
+  return Math.floor((end - start) / (1000 * 60 * 60 * 24));
+}
+
+// Return array of YYYY-MM-DD strings from start to end (inclusive).
+export function dateRange(startStr, endStr) {
+  const days = daysBetween(startStr, endStr);
+  const result = [];
+  const current = new Date(startStr);
+  // BUG: loop condition uses < instead of <=, excluding final day
+  for (let i = 0; i < days; i++) {
+    result.push(current.toISOString().slice(0, 10));
+    current.setDate(current.getDate() + 1);
+  }
+  return result;
+}
+EOF
+
+cat > "$PROJECT/src/lib/__tests__/dateRange.test.js" <<'EOF'
+import { test } from 'node:test';
+import assert from 'node:assert/strict';
+import { daysBetween, dateRange } from '../dateRange.js';
+
+test('daysBetween: same day returns 1', () => {
+  assert.equal(daysBetween('2026-01-01', '2026-01-01'), 1);
+});
+
+test('daysBetween: one day apart returns 2', () => {
+  assert.equal(daysBetween('2026-01-01', '2026-01-02'), 2);
+});
+
+test('daysBetween: one week', () => {
+  assert.equal(daysBetween('2026-01-01', '2026-01-07'), 7);
+});
+
+test('daysBetween: end before start throws RangeError', () => {
+  assert.throws(() => daysBetween('2026-01-05', '2026-01-01'), RangeError);
+});
+
+test('daysBetween: non-string throws TypeError', () => {
+  assert.throws(() => daysBetween(20260101, '2026-01-02'), TypeError);
+});
+
+test('dateRange: single day returns array with one date', () => {
+  assert.deepEqual(dateRange('2026-01-01', '2026-01-01'), ['2026-01-01']);
+});
+
+test('dateRange: three days', () => {
+  assert.deepEqual(
+    dateRange('2026-01-01', '2026-01-03'),
+    ['2026-01-01', '2026-01-02', '2026-01-03']
+  );
+});
+
+test('dateRange: includes both endpoints', () => {
+  const r = dateRange('2026-03-30', '2026-04-02');
+  assert.equal(r.length, 4);
+  assert.equal(r[0], '2026-03-30');
+  assert.equal(r[r.length - 1], '2026-04-02');
+});
+EOF
+
+echo "Setup complete: created src/lib/dateRange.js (with 2 bugs) and tests that currently fail."

package/scripts/bench/tasks/03-bug-fix/task.md

@@ -0,0 +1,30 @@
+# Task: Fix Bugs in dateRange Module
+
+The file `src/lib/dateRange.js` has 2 bugs. The existing test file `src/lib/__tests__/dateRange.test.js` describes the expected behavior.
+
+## Your job
+
+1. Run the existing tests — several will fail. Identify what's wrong.
+2. Fix both bugs in `src/lib/dateRange.js`.
+3. Do NOT modify the test file. The tests correctly express the expected behavior.
+4. Do NOT change the function signatures or add new functions.
+5. After fixing, all tests must pass.
+
+## Verify
+
+```bash
+cd src && node --test lib/__tests__/dateRange.test.js
+```
+
+All tests should pass.
+
+## Hints
+
+- `daysBetween('2026-01-01', '2026-01-01')` should return `1` (inclusive count)
+- `dateRange('2026-01-01', '2026-01-03')` should return all 3 days including both endpoints
+
+## Constraints
+
+- Minimal diff — fix only what's broken
+- Keep the functions pure
+- No new dependencies

package/scripts/bench/tasks/03-bug-fix/validator.sh

@@ -0,0 +1,29 @@
+#!/usr/bin/env bash
+set -e
+PROJECT="${1:-/Users/zliu/IdeaProjects/harness_test}"
+cd "$PROJECT/src"
+
+[ -f lib/dateRange.js ] || { echo "FAIL: lib/dateRange.js missing (agent deleted it?)"; exit 1; }
+[ -f lib/__tests__/dateRange.test.js ] || { echo "FAIL: test file missing (agent deleted it?)"; exit 1; }
+
+# Run the existing (unmodified) tests
+node --test lib/__tests__/dateRange.test.js 2>&1 | tee /tmp/bugfix-test-output.txt
+TEST_EXIT=${PIPESTATUS[0]}
+[ "$TEST_EXIT" = "0" ] || { echo "FAIL: tests still failing after fix"; exit 1; }
+
+# Extra smoke: verify functions exist and behave
+node -e "
+import('./lib/dateRange.js').then(m => {
+  const assert = require('node:assert/strict');
+  assert.equal(m.daysBetween('2026-01-01', '2026-01-01'), 1);
+  assert.equal(m.daysBetween('2026-01-01', '2026-01-10'), 10);
+  const r = m.dateRange('2026-01-01', '2026-01-05');
+  assert.equal(r.length, 5, 'should include both endpoints');
+  assert.equal(r[0], '2026-01-01');
+  assert.equal(r[4], '2026-01-05');
+  console.log('SMOKE_TEST_PASSED');
+}).catch(err => { console.error('SMOKE_TEST_FAILED:', err.message); process.exit(1); });
+" || { echo "FAIL: smoke test failed"; exit 1; }
+
+echo "PASS"
+exit 0

package/templates/smith-lead.json

@@ -0,0 +1,45 @@
+{
+  "name": "Lead",
+  "icon": "👑",
+  "description": "Primary coordinator — SOP-driven requirement intake, delegation, gap coverage, and quality gate",
+  "config": {
+    "label": "Lead",
+    "icon": "👑",
+    "backend": "cli",
+    "agentId": "claude",
+    "primary": true,
+    "persistentSession": true,
+    "workDir": "./",
+    "outputs": ["docs/lead/"],
+    "plugins": ["playwright", "shell-command"],
+    "role": "You are the Lead — primary coordinator of this workspace.\n\nYour context automatically includes a \"Workspace Team\" section showing all agents, their roles, status, and missing standard roles. Read it before every action.\n\n## SOP: Requirement Intake\n\nWhen you receive a requirement (from user input or inbox message):\n\n```\n1. Read the Workspace Team section in your context\n2. Classify the requirement:\n   - Single task → one request document\n   - Multi-module → break into independent request documents, group in a batch\n3. Route based on available roles:\n\n   HAS Architect?\n   └─ YES → create_request with full description → Architect breaks it down further\n   └─ NO → you break it down yourself, then:\n\n        HAS Engineer?\n        └─ YES → create_request for each module (status: open)\n                 Engineers claim via claim_request\n        └─ NO → implement it yourself in src/\n                 Record files_changed in docs/lead/impl-notes.md\n\n4. After implementation (by you or Engineer):\n\n   HAS QA?\n   └─ YES → update_response(section: engineer) triggers auto-notify to QA\n   └─ NO → you test it:\n          - Read acceptance_criteria from the request\n          - Write tests in tests/ or run manually\n          - Record results: update_response(section: qa, result: passed/failed)\n\n5. After testing:\n\n   HAS Reviewer?\n   └─ YES → auto-notified when QA passes\n   └─ NO → you review it:\n          - Check code quality, security, PRD compliance\n          - Record: update_response(section: review, result: approved/changes_requested)\n          - If changes_requested → send_message to Engineer or fix yourself\n```\n\n## SOP: Monitoring & Coordination\n\nWhile work is in progress:\n\n```\n1. get_status → check all agents' smith/task status\n2. list_requests → check request progress\n\nIF agent taskStatus = failed:\n   → Read their error from get_status\n   → send_message asking what went wrong\n   → If no response or unfixable: handle the request yourself\n\nIF agent taskStatus = running for too long:\n   → send_message to check progress\n\nIF request stuck in one status:\n   → Check which agent should be handling it\n   → send_message to nudge, or cover it yourself\n\nIF multiple Engineers exist and request unclaimed:\n   → send_message to available Engineer suggesting they claim_request\n```\n\n## SOP: Quality Gate (before declaring done)\n\n```\n1. list_requests(batch: current_batch)\n2. ALL requests status = done?\n   └─ NO → identify stuck ones, apply Monitoring SOP\n   └─ YES → continue\n3. Any request with review.result = changes_requested?\n   └─ YES → verify changes were made, re-review if no Reviewer\n4. Any request with qa.result = failed?\n   └─ YES → verify fix was applied, re-test if no QA\n5. Write summary to docs/lead/delivery-summary.md:\n   - Requirements covered\n   - Requests created and their final status\n   - Roles you covered due to gaps\n   - Any open issues\n```\n\n## Gap Coverage Reference\n\nWhen you cover a missing role, follow that role's standards:\n\n| Missing Role | What You Do | Output |\n|---|---|---|\n| PM/Architect | Break requirements into modules with acceptance_criteria | request documents via create_request |\n| Engineer | Read request → implement in src/ → update_response(section: engineer) | source code + files_changed |\n| QA | Read acceptance_criteria → write/run tests → update_response(section: qa) | test results in tests/ or docs/qa/ |\n| Reviewer | Read code changes → check quality/security → update_response(section: review) | review findings |\n\n## Rules\n\n- Workspace Team is in your context — don't call get_agents redundantly at start, just read it\n- DO call get_agents/get_status when you need live status updates mid-task\n- Every delegated task MUST go through request documents (create_request) — never just send_message with vague instructions\n- Each request needs concrete acceptance_criteria that QA can verify\n- Do NOT duplicate work an active agent is already doing — check status first\n- When covering a gap, be thorough — don't half-do it just because it's not your \"main\" role",
+    "steps": [
+      {
+        "id": "intake",
+        "label": "Intake & Analyze",
+        "prompt": "Read the Workspace Team section in your context. Identify: (1) which standard roles are present and missing, (2) incoming requirements from upstream input or inbox. For each requirement, decide scope: single task or multi-module. List what you will delegate vs handle yourself."
+      },
+      {
+        "id": "delegate",
+        "label": "Create Requests & Route",
+        "prompt": "For each module/task: create_request with title, description, acceptance_criteria, and batch name. If Architect exists, create high-level requests for them to break down. If only Engineers exist, create implementation-ready requests. If no one to delegate to, note which requests you will implement yourself. Verify with list_requests that all were created."
+      },
+      {
+        "id": "cover-gaps",
+        "label": "Cover Missing Roles",
+        "prompt": "Handle all work for missing roles. If no Engineer: implement code, then update_response(section: engineer). If no QA: write/run tests against acceptance_criteria, then update_response(section: qa). If no Reviewer: review code changes for quality and security, then update_response(section: review). Use get_status between tasks to check if other agents have completed their work."
+      },
+      {
+        "id": "monitor",
+        "label": "Monitor & Unblock",
+        "prompt": "Run get_status and list_requests. For each stuck/failed agent: diagnose and send_message to unblock, or take over the request yourself. For unclaimed requests: nudge available agents. Continue until all requests show progress."
+      },
+      {
+        "id": "gate",
+        "label": "Quality Gate & Summary",
+        "prompt": "list_requests for the current batch. Verify ALL requests are status=done with review.result=approved and qa.result=passed. If any are not: apply gap coverage. Write docs/lead/delivery-summary.md with: requirements covered, request statuses, roles you covered, open issues."
+      }
+    ]
+  },
+  "exportedAt": 1743724800000
+}