Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion autoresearch/eval-publish.ts
Original file line number Diff line number Diff line change
Expand Up @@ -78,7 +78,7 @@ function judge(criteria: JudgeCriteria, output: string): boolean {
}

function runCommand(cmd: string, timeout = 30000): string {
const localCmd = cmd.replace(/^opencli /, `node dist/main.js `);
const localCmd = cmd.replace(/^opencli /, `node dist/src/main.js `);
try {
return execSync(localCmd, {
cwd: PROJECT_ROOT,
Expand Down
2 changes: 1 addition & 1 deletion autoresearch/eval-save.ts
Original file line number Diff line number Diff line change
Expand Up @@ -83,7 +83,7 @@ const PROJECT_ROOT = join(__dirname, '..');
/** Run a command, using local dist/main.js instead of global opencli for consistency */
function runCommand(cmd: string, timeout = 30000): string {
// Use local build so tests always run against the current source
const localCmd = cmd.replace(/^opencli /, `node dist/main.js `);
const localCmd = cmd.replace(/^opencli /, `node dist/src/main.js `);
try {
return execSync(localCmd, {
cwd: PROJECT_ROOT,
Expand Down
203 changes: 203 additions & 0 deletions skills/opencli-repair/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,203 @@
---
name: opencli-repair
description: Diagnose and fix broken OpenCLI adapters when websites change. Use when an opencli command fails with SELECTOR, EMPTY_RESULT, API_ERROR, or PAGE_CHANGED errors. Reads structured diagnostic output and uses browser automation to discover what changed and patch the adapter.
allowed-tools: Bash(opencli:*), Read, Edit, Write
---

# OpenCLI Repair — AI-Driven Adapter Self-Repair

When an adapter breaks because a website changed its DOM, API, or auth flow, use this skill to diagnose the failure and patch the adapter.

## Prerequisites

```bash
opencli doctor # Verify extension + daemon connectivity
```

## When to Use This Skill

Use when `opencli <site> <command>` fails with errors like:
- **SELECTOR** — element not found (DOM changed)
- **EMPTY_RESULT** — no data returned (API response changed)
- **API_ERROR** / **NETWORK** — endpoint moved or broke
- **PAGE_CHANGED** — page structure no longer matches
- **COMMAND_EXEC** — runtime error in adapter logic
- **TIMEOUT** — page loads differently, adapter waits for wrong thing

## Step 1: Collect Diagnostic Context

Run the failing command with diagnostic mode enabled:

```bash
OPENCLI_DIAGNOSTIC=1 opencli <site> <command> [args...] 2>diagnostic.json
```

This outputs a `RepairContext` JSON between `___OPENCLI_DIAGNOSTIC___` markers in stderr:

```json
{
"error": {
"code": "SELECTOR",
"message": "Could not find element: .old-selector",
"hint": "The page UI may have changed."
},
"adapter": {
"site": "example",
"command": "example/search",
"sourcePath": "/path/to/clis/example/search.ts",
"source": "// full adapter source code"
},
"page": {
"url": "https://example.com/search",
"snapshot": "// DOM snapshot with [N] indices",
"networkRequests": [],
"consoleErrors": []
},
"timestamp": "2025-01-01T00:00:00.000Z"
}
```

**Parse it:**
```bash
# Extract JSON between markers from stderr output
cat diagnostic.json | sed -n '/___OPENCLI_DIAGNOSTIC___/{n;p;}'
```

## Step 2: Analyze the Failure

Read the diagnostic context and the adapter source. Classify the root cause:

| Error Code | Likely Cause | Repair Strategy |
|-----------|-------------|-----------------|
| SELECTOR | DOM restructured, class/id renamed | Explore current DOM → find new selector |
| EMPTY_RESULT | API response schema changed, or data moved | Check network → find new response path |
| API_ERROR | Endpoint URL changed, new params required | Discover new API via network intercept |
| AUTH_REQUIRED | Login flow changed, cookies expired | Walk login flow with operate |
| TIMEOUT | Page loads differently, spinner/lazy-load | Add/update wait conditions |
| PAGE_CHANGED | Major redesign | May need full adapter rewrite |

**Key questions to answer:**
1. What is the adapter trying to do? (Read the `source` field)
2. What did the page look like when it failed? (Read the `snapshot` field)
3. What network requests happened? (Read `networkRequests`)
4. What's the gap between what the adapter expects and what the page provides?

## Step 3: Explore the Current Website

Use `opencli operate` to inspect the live website. **Never use the broken adapter** — it will just fail again.

### DOM changed (SELECTOR errors)

```bash
# Open the page and inspect current DOM
opencli operate open https://example.com/target-page && opencli operate state

# Look for elements that match the adapter's intent
# Compare the snapshot with what the adapter expects
```

### API changed (API_ERROR, EMPTY_RESULT)

```bash
# Open page with network interceptor, then trigger the action manually
opencli operate open https://example.com/target-page && opencli operate state

# Interact to trigger API calls
opencli operate click <N> && opencli operate network

# Inspect specific API response
opencli operate network --detail <index>
```

### Auth changed (AUTH_REQUIRED)

```bash
# Check current auth state
opencli operate open https://example.com && opencli operate state

# If login page: inspect the login form
opencli operate state # Look for login form fields
```

## Step 4: Patch the Adapter

Read the adapter source file and make targeted fixes:

```bash
# Read the adapter
cat <sourcePath from diagnostic>
```

### Common Fixes

**Selector update:**
```typescript
// Before: page.evaluate('document.querySelector(".old-class")...')
// After: page.evaluate('document.querySelector(".new-class")...')
```

**API endpoint change:**
```typescript
// Before: const resp = await page.evaluate(`fetch('/api/v1/old-endpoint')...`)
// After: const resp = await page.evaluate(`fetch('/api/v2/new-endpoint')...`)
```

**Response schema change:**
```typescript
// Before: const items = data.results
// After: const items = data.data.items // API now nests under "data"
```

**Wait condition update:**
```typescript
// Before: await page.waitForSelector('.loading-spinner', { hidden: true })
// After: await page.waitForSelector('[data-loaded="true"]')
```

### Rules for Patching

1. **Make minimal changes** — fix only what's broken, don't refactor
2. **Keep the same output structure** — `columns` and return format must stay compatible
3. **Prefer API over DOM scraping** — if you discover a JSON API during exploration, switch to it
4. **Use `@jackwener/opencli/*` imports only** — never add third-party package imports
5. **Test after patching** — run the command again to verify

## Step 5: Verify the Fix

```bash
# Run the command normally (without diagnostic mode)
opencli <site> <command> [args...]
```

If it still fails, go back to Step 3 and explore further. If the website has fundamentally changed (major redesign, removed feature), report that the adapter needs a full rewrite.

## When to Give Up

Not all failures are repairable with a quick patch:

- **Site requires CAPTCHA** — can't automate this
- **Feature completely removed** — the data no longer exists
- **Major redesign** — needs full adapter rewrite via `opencli-explorer` skill
- **Rate limited / IP blocked** — not an adapter issue

In these cases, clearly communicate the situation to the user rather than making futile patches.

## Example Repair Session

```
1. User runs: opencli zhihu hot
→ Fails: SELECTOR "Could not find element: .HotList-item"

2. AI runs: OPENCLI_DIAGNOSTIC=1 opencli zhihu hot 2>diag.json
→ Gets RepairContext with DOM snapshot showing page loaded

3. AI reads diagnostic: snapshot shows the page loaded but uses ".HotItem" instead of ".HotList-item"

4. AI explores: opencli operate open https://www.zhihu.com/hot && opencli operate state
→ Confirms new class name ".HotItem" with child ".HotItem-content"

5. AI patches: Edit clis/zhihu/hot.ts — replace ".HotList-item" with ".HotItem"

6. AI verifies: opencli zhihu hot
→ Success: returns hot topics
```
100 changes: 100 additions & 0 deletions src/diagnostic.test.ts
Original file line number Diff line number Diff line change
@@ -0,0 +1,100 @@
import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
import { buildRepairContext, isDiagnosticEnabled, emitDiagnostic, type RepairContext } from './diagnostic.js';
import { CliError, SelectorError, CommandExecutionError } from './errors.js';
import type { InternalCliCommand } from './registry.js';

function makeCmd(overrides: Partial<InternalCliCommand> = {}): InternalCliCommand {
return {
site: 'test-site',
name: 'test-cmd',
description: 'test',
args: [],
...overrides,
} as InternalCliCommand;
}

describe('isDiagnosticEnabled', () => {
const origEnv = process.env.OPENCLI_DIAGNOSTIC;
afterEach(() => {
if (origEnv === undefined) delete process.env.OPENCLI_DIAGNOSTIC;
else process.env.OPENCLI_DIAGNOSTIC = origEnv;
});

it('returns false when env not set', () => {
delete process.env.OPENCLI_DIAGNOSTIC;
expect(isDiagnosticEnabled()).toBe(false);
});

it('returns true when env is "1"', () => {
process.env.OPENCLI_DIAGNOSTIC = '1';
expect(isDiagnosticEnabled()).toBe(true);
});

it('returns false for other values', () => {
process.env.OPENCLI_DIAGNOSTIC = 'true';
expect(isDiagnosticEnabled()).toBe(false);
});
});

describe('buildRepairContext', () => {
it('captures CliError fields', () => {
const err = new SelectorError('.missing-element', 'Element removed');
const ctx = buildRepairContext(err, makeCmd());

expect(ctx.error.code).toBe('SELECTOR');
expect(ctx.error.message).toContain('.missing-element');
expect(ctx.error.hint).toBe('Element removed');
expect(ctx.error.stack).toBeDefined();
expect(ctx.adapter.site).toBe('test-site');
expect(ctx.adapter.command).toBe('test-site/test-cmd');
expect(ctx.timestamp).toMatch(/^\d{4}-\d{2}-\d{2}T/);
});

it('handles non-CliError errors', () => {
const err = new TypeError('Cannot read property "x" of undefined');
const ctx = buildRepairContext(err, makeCmd());

expect(ctx.error.code).toBe('UNKNOWN');
expect(ctx.error.message).toContain('Cannot read property');
expect(ctx.error.hint).toBeUndefined();
});

it('includes page state when provided', () => {
const pageState: RepairContext['page'] = {
url: 'https://example.com/page',
snapshot: '<div>...</div>',
networkRequests: [{ url: '/api/data', status: 200 }],
consoleErrors: ['Uncaught TypeError'],
};
const ctx = buildRepairContext(new CommandExecutionError('boom'), makeCmd(), pageState);

expect(ctx.page).toEqual(pageState);
});

it('omits page when not provided', () => {
const ctx = buildRepairContext(new Error('boom'), makeCmd());
expect(ctx.page).toBeUndefined();
});
});

describe('emitDiagnostic', () => {
it('writes delimited JSON to stderr', () => {
const writeSpy = vi.spyOn(process.stderr, 'write').mockReturnValue(true);

const ctx = buildRepairContext(new CommandExecutionError('test error'), makeCmd());
emitDiagnostic(ctx);

const output = writeSpy.mock.calls.map(c => c[0]).join('');
expect(output).toContain('___OPENCLI_DIAGNOSTIC___');
expect(output).toContain('"code":"COMMAND_EXEC"');
expect(output).toContain('"message":"test error"');

// Verify JSON is parseable between markers
const match = output.match(/___OPENCLI_DIAGNOSTIC___\n(.*)\n___OPENCLI_DIAGNOSTIC___/);
expect(match).toBeTruthy();
const parsed = JSON.parse(match![1]);
expect(parsed.error.code).toBe('COMMAND_EXEC');

writeSpy.mockRestore();
});
});
Loading
Loading