Files
jackwener__opencli/clis/gemini/image.js
jakevin dc724262f2 feat: agent-native retrospective — analyze / verify guards / fixture content checks (#1133)
* feat: agent-native retrospective — analyze / verify guards / fixture content checks

Post-mortem on slow 1point3acres + 51job adapter sessions, consolidated
into one PR. Scope is "reduce uncertainty and catch silent failures"
— the two things that sink agent success rate on first-time adapters.

Changes:
- `browser analyze <url>` — one command returns pattern (A/B/C/D),
  anti-bot vendor (Aliyun/Cloudflare/Akamai/Geetest), nearest adapter,
  and a single-sentence recommended_next_step. Replaces the three-step
  open/wait/network recon loop when it can reach a confident verdict.
- `browser wait xhr <regex>` — poll for a specific XHR URL instead of
  blind `wait time N`, so SPA data-arrival barriers are deterministic.
- Fixture `mustNotContain` / `mustBeTruthy` — catch two silent-failure
  modes `notEmpty` misses: content contamination (sibling DOM bleed)
  and `|| 0` / `|| false` fallbacks.
- `browser verify` post-success site-memory check + `--strict-memory`
  — verify-green no longer hides the case where `~/.opencli/sites/`
  was never written back. Memory only materializes if authors write it.
- CI: guard that committed `cli-manifest.json` matches a fresh build.
  Main was already drifted (#1118 left stale ordering + a missing arg);
  this PR regenerates the manifest and will catch the next drift.

Docs (opencli-adapter-author + opencli-autofix skills):
- `success-rate-pitfalls.md` — 10 concrete silent-failure scenarios
  seen in real adapter sessions, each with defense via fixture /
  adapter patterns.
- `autofix` gains discipline rule #6: verify pattern failure means
  tighten the adapter, never loosen the fixture.
- `site-recon.md` leads with `browser analyze`; `api-discovery.md`
  adds a §0 covering WAF vendor detection and cross-subdomain CORS
  (the two gotchas that burned the 51job session).
- `wait time 3` → `wait time 2`, with `wait xhr` as the robust choice.

* fix: make output-dir defaults host-independent in manifest

Three adapters (chatgpt/image, gemini/image, instagram/download) baked
`path.join(os.homedir(), ...)` into the `default` field of their args.
The committed manifest therefore carried my personal `/Users/jakevin/...`
paths — which agents running on a different host saw as surprising
defaults. The drift guard I just added to CI caught it on the first run.

Runtime behavior is unchanged: each adapter still falls back to
`path.join(os.homedir(), …)` inside `func` when the kwarg is absent.
Only the displayed / registered default becomes a tilde-path.

* fix(cli): enforce strict-memory without fixture

* fix(browser): harden analyze and xhr guards

* fix(browser): fallback to interceptor buffer
2026-04-22 01:59:19 +08:00

106 lines
4.5 KiB
JavaScript

import * as os from 'node:os';
import * as path from 'node:path';
import { cli, Strategy } from '@jackwener/opencli/registry';
import { saveBase64ToFile } from '@jackwener/opencli/utils';
import { GEMINI_DOMAIN, exportGeminiImages, getGeminiVisibleImageUrls, sendGeminiMessage, startNewGeminiChat, waitForGeminiImages } from './utils.js';
function extFromMime(mime) {
if (mime.includes('png'))
return '.png';
if (mime.includes('webp'))
return '.webp';
if (mime.includes('gif'))
return '.gif';
return '.jpg';
}
function normalizeBooleanFlag(value) {
if (typeof value === 'boolean')
return value;
const normalized = String(value ?? '').trim().toLowerCase();
return normalized === 'true' || normalized === '1' || normalized === 'yes' || normalized === 'on';
}
function displayPath(filePath) {
const home = os.homedir();
return filePath.startsWith(home) ? `~${filePath.slice(home.length)}` : filePath;
}
function buildImagePrompt(prompt, options) {
const extras = [];
if (options.ratio)
extras.push(`aspect ratio ${options.ratio}`);
if (options.style)
extras.push(`style ${options.style}`);
if (extras.length === 0)
return prompt;
return `${prompt}
Image requirements: ${extras.join(', ')}.`;
}
function normalizeRatio(value) {
const normalized = value.trim();
const allowed = new Set(['1:1', '16:9', '9:16', '4:3', '3:4', '3:2', '2:3']);
return allowed.has(normalized) ? normalized : '1:1';
}
async function currentGeminiLink(page) {
const url = await page.evaluate('window.location.href').catch(() => '');
return typeof url === 'string' && url ? url : 'https://gemini.google.com/app';
}
export const imageCommand = cli({
site: 'gemini',
name: 'image',
description: 'Generate images with Gemini web and save them locally',
domain: GEMINI_DOMAIN,
strategy: Strategy.COOKIE,
browser: true,
navigateBefore: false,
defaultFormat: 'plain',
timeoutSeconds: 240,
args: [
{ name: 'prompt', positional: true, required: true, help: 'Image prompt to send to Gemini' },
{ name: 'rt', default: '1:1', help: 'Ratio shorthand for aspect ratio (1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3)' },
{ name: 'st', default: '', help: 'Style shorthand, e.g. anime, icon, watercolor' },
{ name: 'op', default: '~/tmp/gemini-images', help: 'Output directory shorthand' },
{ name: 'sd', type: 'boolean', default: false, help: 'Skip download shorthand; only show Gemini page link' },
],
columns: ['status', 'file', 'link'],
func: async (page, kwargs) => {
const prompt = kwargs.prompt;
const ratio = normalizeRatio(String(kwargs.rt ?? '1:1'));
const style = String(kwargs.st ?? '').trim();
const outputDir = kwargs.op || path.join(os.homedir(), 'tmp', 'gemini-images');
const timeout = 120;
const startFresh = true;
const skipDownloadRaw = kwargs.sd;
const skipDownload = skipDownloadRaw === '' || skipDownloadRaw === true || normalizeBooleanFlag(skipDownloadRaw);
const effectivePrompt = buildImagePrompt(prompt, {
ratio,
style: style || undefined,
});
if (startFresh)
await startNewGeminiChat(page);
const beforeUrls = await getGeminiVisibleImageUrls(page);
await sendGeminiMessage(page, effectivePrompt);
const urls = await waitForGeminiImages(page, beforeUrls, timeout);
const link = await currentGeminiLink(page);
if (!urls.length) {
return [{ status: '⚠️ no-images', file: '📁 -', link: `🔗 ${link}` }];
}
if (skipDownload) {
return [{ status: '🎨 generated', file: '📁 -', link: `🔗 ${link}` }];
}
const assets = await exportGeminiImages(page, urls);
if (!assets.length) {
return [{ status: '⚠️ export-failed', file: '📁 -', link: `🔗 ${link}` }];
}
const stamp = Date.now();
const results = [];
for (let index = 0; index < assets.length; index += 1) {
const asset = assets[index];
const base64 = asset.dataUrl.replace(/^data:[^;]+;base64,/, '');
const suffix = assets.length > 1 ? `_${index + 1}` : '';
const filePath = path.join(outputDir, `gemini_${stamp}${suffix}${extFromMime(asset.mimeType)}`);
await saveBase64ToFile(base64, filePath);
results.push({ status: '✅ saved', file: `📁 ${displayPath(filePath)}`, link: `🔗 ${link}` });
}
return results;
},
});