mirror of
https://github.com/garrytan/gstack.git
synced 2026-09-09 06:28:59 +02:00
* feat(aside): browser-driver contract, cookbook, research and fallback resolvers
{{ASIDE_SETUP}} (readiness probe + ten rules for driving the user's real browser), {{ASIDE_COOKBOOK}} (script shapes verified live against Aside CLI 1.26: one flow per aside repl script, CDP console hook before navigation, evidence lines, session-directory artifact handoff, GSTACK_STEP_OK sentinel), {{ASIDE_RESEARCH}} (research through aside exec, WebSearch when Aside is absent, knowledge otherwise) and {{BROWSE_FALLBACK}} (the fifteen-row Aside-step to $B-command table plus the rules that differ, so every browsing skill keeps working on gstack's own headless browser). test/aside-driver.test.ts pins the sentences and asserts every browsing skill carries the Aside block followed by the fallback; test/helpers/aside-available.ts is the shared live-Aside probe.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* feat(render): Aside-first local-HTML renderer with the bundled browser as fallback
lib/aside-render.ts serves the HTML's directory on loopback (Aside refuses file:// URLs), opens it with waitUntil load, prints through CDP Page.printToPDF so tagged output, outlines, header/footer templates and page numbers survive, emulates device metrics for sized screenshots, and writes in-page evaluations to files; when Aside is absent it runs the same spec through the browse daemon (newtab, load, js, pdf, screenshot, closetab) and reports ENGINE=aside|browse. bin/gstack-render.ts is the CLI skill templates call. lib/claude-bin.ts and lib/error-handling.ts become the canonical copies (browse/src re-exports them).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(browse): /browse drives Aside first, with the $B reference behind the fallback
Contract, cookbook, mode choice (aside repl by default, aside exec for reading), report format, the fallback section, and the full command reference carved on demand.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(qa): /qa and /qa-only drive Aside, fall back to $B
QA_METHODOLOGY runs every phase as Aside scripts (orient, explore, document, re-test, mobile viewport via CDP emulation, links via HEAD fetch); the authenticate phase is 'you are already signed in'; a 13th rule requires consent before mutating actions on non-local targets; the fallback section translates each step onto $B. The qa E2E tests run on whichever engine is present.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(design): design-review, design-consultation, design-shotgun, plan-design-review, design-html drive Aside
Design-system extraction is one script printing FONTS/COLORS/HEADINGS/TOUCH_TARGETS/NAV; competitor research confirms the exact URLs before opening them in the real browser and runs on the bundled browser when Aside is absent; design-html's viewport screenshots, sketches and comparison boards render through gstack-render.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(deploy): benchmark, canary, land-and-deploy Step 7, devex-review drive Aside
One aside repl script per page prints NAV/PAINT/LCP/RESOURCES/SCRIPTS/CSS/SUMMARY (benchmark), CONSOLE_ERRORS/NAV/TEXT + screenshot (canary, re-run every 60s), and the post-deploy check reads responseStatus from the navigation entry; each carries the $B fallback.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* feat(third-party-actions): Aside is the recommended driver; gstack's visible browser stays the fallback
The readiness probe is lifted from {{ASIDE_SETUP}} at gen time (byte-identity pinned) and rule 3 points at browse/SKILL.md for how to drive; the consent question offers Aside first and gstack's own visible browser (handoff/resume for sign-in) as the fallback, as v1.72 framed it.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(scrape): /scrape reads pages through Aside; the browser-skills runtime rides the fallback
Look-then-extract scripts build the JSON inside the page and print it between JSON_START/JSON_END; aside exec for fuzzy intents; on the $B fallback the browser-skills match/prototype flow and /skillify apply as before.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(make-pdf): print through Aside first, the bundled browser otherwise
asideClient.ts replaces the direct $B client with one render() call per PDF (the exact option mapping the browse pdf command had: paper, margins, header/footer/page numbers, tagged, outline, printBackground, preferCSSPageSize, Paged.js wait); the diagram pre-pass, oversized-image downscale and DOCX rasters each run as one render script with per-fence try/catch; exit 4 now means no browser is available and names both remedies; $P setup reports which engine it found. The e2e gates run on whichever engine is present, so the Linux lane exercises the fallback.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(diagram): the triplet is one gstack-render call
SVG, PNG and excalidraw from one invocation over the content-addressed bundle staged under /tmp/gstack-render; every diagram type gets an excalidraw export; gstack-render picks the engine and prints ENGINE=; the diagram E2E gates on either engine.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* feat(research): web research runs in Aside first, WebSearch second
The planning, review, design, security and investigate skills research through {{ASIDE_RESEARCH}}; WebSearch stays in allowed-tools as the fallback; testing.ts's bootstrap step follows; skeleton ceilings ratcheted for the research block.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* feat(setup,gen-skill-docs): prune renders of skills that no longer exist
setup gains _prune_stale_generated for every host tree and the doc generator removes gstack-* output dirs it did not write, so a skill removed from the source tree can never linger in an install.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* test: registries, budgets and suite reconciled for Aside-first with the $B fallback
Touchfiles + E2E tiers gain the Aside keys, coverage matrix and eval baselines updated, size budget re-baselined to parity-baseline-v1.80.0.0.json (the contract plus fallback ride in every browsing skill), parity ceilings ratcheted with measured values, LLM-judge prompts and the E2E fixtures speak Aside-first, browse-fallback.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: Aside first, gstack browser fallback
README, BROWSER.md, docs/, CONTRIBUTING, CLAUDE.md, ARCHITECTURE, AGENTS.md, TODOS and the root router describe the one product story: Aside is the browser gstack drives first; the bundled headless browser is the automatic fallback (Linux, Windows, app closed) where cookie import, GStack Browser, pair-agent and browser-skills still apply.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* chore: regenerate SKILL.md docs, llms.txt, agents digest, ship goldens, context-budget fixture
bun run gen:skill-docs over the templates; goldens re-rendered; context-budget ceilings recaptured.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* v1.80.0.0: Aside is the browser gstack drives first; the bundled browser is the fallback
MINOR: new capability across ten skills, the renderer and research; nothing removed. CHANGELOG release summary + itemized changes; VERSION 1.80.0.0; package.json 1.80.0.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs(todos): file non-Claude host ownership-gate and version-heading pin follow-ups
Two follow-ups from the /plan-ceo-review + /plan-eng-review pass on merging
PR #2804 with main's v1.80.0.0 ownership gate: bring the Codex/Factory/
OpenCode/Cursor/Kiro copy loops and the stale-render prune under the
.gstack-owned marker rule, and a free test pinning that the CHANGELOG top
heading equals VERSION (the collision that git cannot see).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: pre-landing review fixes for the Aside-first branch
Review army + adversarial passes (Claude and Codex) on the merged branch:
setup
- _prune_stale_generated scans the host dirs too (the generator already
removed the render before setup ran, so the host branch was dead), skips
symlinks in the render tree (rm -rf on a slash-terminated link empties its
target), removes a host symlink only when it resolves into gstack, cleans a
bannered real dir through _cleanup_weak_dir, recognizes frontmatter-renamed
skills, and logs through log. The always-run codex render passes every host
dir that may link to it.
- NEEDS_BUILD checks all three binaries (with $_EXE) and lib/ sources; the
browser hint and the bootstrap summary honor GSTACK_SKIP_ASIDE, treat a
requested skip as a request, and derive one skill list.
lib/aside-render.ts + bin/gstack-render.ts
- The loopback server carries a per-render secret path, checks containment on
the real path (symlink escapes are 403), and rejects malformed encoding.
- Inline eval results are one base64 line, so page text cannot forge
ASIDE_DIR= or the sentinel; the last ASIDE_DIR wins.
- runProc escalates SIGTERM to SIGKILL, bounds every wait, and clears every
timer (an uncleared one kept gstack-render alive after printing OK).
- renderTmpDir refuses a shared /tmp name owned by someone else; the work dir
and server are created inside try; goto's budget follows the render budget.
- probeAside classifies a present-but-failing CLI as ASIDE_NOT_RUNNING like
the skills' bash probe; render() retries on gstack's own browser when Aside
could not start or its private CDP bridge is gone (never on a page error
or a timeout of a running script); the CLI reports the engine that actually
rendered, exits 0 on --help, rejects non-numeric flags, documents
--wait-timeout, fences EVAL/PAGE_ERRORS as untrusted content, and names the
daemon's cookie-import JS lock remedy.
- The browse path passes --scale only when asked (a scale change rebuilds
the daemon context) and restores the viewport after a sized screenshot.
resolvers / templates
- The bash probe honors GSTACK_SKIP_ASIDE and has a perl deadline on stock
macOS; .local is no longer LOCAL (mDNS); same-origin filters compare parsed
origins; link status is HEAD-checked only on LOCAL targets; every
aside exec goes through the receipted _aside_exec prelude
({{ASIDE_EXEC_PRELUDE}}), including nine template blocks that called it
bare; the design sketch and diagram staging use private directories.
- The generator prunes only bannered renders and never a host whose
generation failed.
Docs, stale comments and dead code cleaned; goldens re-rendered; tests
updated and added for every behavior above.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* test: coverage for the render CLI, setup rebuild check, make-pdf exit codes, and prose $B spans
New free tests from the ship coverage audit: test/gstack-render-cli.test.ts
(argv guards, --help, output contract with a fake daemon, failure and
serve-root paths, no-browser case, prompt exit), test/setup-needs-build.test.ts
(every binary and source set flips NEEDS_BUILD, Windows suffixes),
make-pdf/test/cli-exit-codes.test.ts and setup-smoke.test.ts (error to exit
code mapping, runSetup stages, renderPdf's engine), and prose-span cases for
extractBrowseCommands in test/skill-parser.test.ts.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: CHANGELOG and TODOS cover the review fixes (v1.81.0.0)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: sync project docs with the v1.81.0.0 review fixes
BROWSER.md, ARCHITECTURE.md, CONTRIBUTING.md, README.md, CLAUDE.md,
docs/TESTING_INTERNALS.md and docs/PROJECT_STRUCTURE.md now describe the
shipped renderer and setup: the loopback render server's per-render secret
path and real-path containment, ENGINE= naming the engine that actually
rendered (mid-run retry on gstack's own browser), EVAL/PAGE_ERRORS fenced as
untrusted content, --wait-timeout and the CLI's argv guards, the receipted
_aside_exec prelude ({{ASIDE_EXEC_PRELUDE}} in the placeholder table), the
LOCAL host rule without .local, LOCAL-only HEAD checks in the links script,
GSTACK_SKIP_ASIDE across probe/renderer/setup, the ownership-gated
retired-skill prune, the widened NEEDS_BUILD check, and the new free tests
(gstack-render-cli, setup-prune-stale-generated, setup-browser-hint,
setup-needs-build, make-pdf cli-exit-codes and setup-smoke).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: CHANGELOG states the precise mid-run retry rule
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(test): skill-e2e-bws slices the $B setup block from the Browser fallback section
browse/SKILL.md no longer has '## SETUP' / '## Core QA Patterns' (Aside is the
primary driver; the $B block moved under 'Browser fallback'), so the gate test
sliced an empty block and handed the agent nothing to run. Anchor on
'### Find the `$B` binary' up to the next heading. 7/7 pass.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(test): gate POSIX-only fixtures off Windows
windows-free-tests: the gstack-render CLI tests drive a shebang fake browse
that CreateProcess cannot exec, and two NEEDS_BUILD cases assert an execute
bit and a bare-name miss that MSYS bash does not have (test -x ignores mode
bits and resolves design -> design.exe). Those describes and cases now
self-skip on win32; argument guards, --help, the no-browser case, and every
other rebuild-check case still run there.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(render): runProc waits for the exit code until the kill deadline; newtab retries once on a cold daemon
A process whose pipes have reached EOF is exiting, but runProc gave the exit
code only five seconds to arrive and then returned null, which run() reports
as a failed command. Under CI's six-shard load one such render failed with the
artifact already written. The SIGTERM/SIGKILL timers already bound the wait,
so the exit race now runs to the kill deadline.
The first CLI call auto-starts the browse daemon; on a cold start it can
answer 'Unable to connect' once while the server is still coming up. That
single case is retried after 1.5s; every other newtab failure is not.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* test(aside-render): warm the daemon before live fallback cases; failures name the render error
- Live fallback cases run 'goto about:blank' up to twice before asserting and
skip (never fail) when the daemon cannot come up.
- expectOk() puts r.error and the browse transcript into the assertion so a
failed render is diagnosable from the CI log.
- The argv-contract cases dump the fake's log on a miss.
- File default timeout is 30s: the subject is the CLI contract, not latency.
- Two cases pin the cold-daemon newtab retry and that other errors are not
retried.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: CHANGELOG notes the cold-start tolerance of the bundled-browser renderer
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Sina <sdroid674+github@gmail.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
303 lines
14 KiB
TypeScript
303 lines
14 KiB
TypeScript
/**
|
|
* Live E2E for the Aside-driven browsing skills — periodic tier.
|
|
*
|
|
* Aside is the primary browser (the user's real sessions); gstack's own
|
|
* headless browser is the fallback. These cases only run with a live Aside.
|
|
* These tests hand a real `claude -p` session the regenerated skill docs and a
|
|
* localhost fixture page, and check that the agent actually drove Aside the way
|
|
* the {{ASIDE_SETUP}} contract (scripts/resolvers/aside.ts) says to: `aside repl`
|
|
* scripts that print labelled evidence and the GSTACK_STEP_OK sentinel, and
|
|
* artifacts copied out of the printed ASIDE_DIR. "Actually drove" means the
|
|
* sentinel appeared on its own line in a BASH tool_result — the skill doc the
|
|
* agent Reads also contains the token, so Read results are excluded and the
|
|
* tokens are line-anchored (Read output is line-numbered; the cookbook wraps
|
|
* them in console.log("…")).
|
|
*
|
|
* The skill docs come from the tree unless GSTACK_E2E_DOCS_ROOT points at a
|
|
* `bun run gen:skill-docs --out-dir <dir>` render (mid-refactor the tree's
|
|
* generated SKILL.md may be stale); either way a copied doc that is not
|
|
* Aside-native fails fast with a "regenerate" message instead of a confusing
|
|
* "agent did not drive Aside".
|
|
*
|
|
* External service (CLAUDE.md tiering rule 3) → periodic. The whole file
|
|
* self-skips — never fails — when EVALS_TIER is not 'periodic' or when
|
|
* `asideAvailable()` is false (no `aside` on PATH, app not running, or
|
|
* GSTACK_SKIP_ASIDE=1). CI runners have no Aside, so this only runs on a dev
|
|
* Mac with the app open:
|
|
*
|
|
* EVALS=1 EVALS_TIER=periodic EVALS_ALL=1 bun test test/skill-e2e-aside.test.ts
|
|
*
|
|
* Cost: ~$1-3 per run (five sessions, 30-40 turns each). Each session gets
|
|
* its own tmp workdir with only the skill's runtime files copied in and
|
|
* cleans it up in a finally.
|
|
*/
|
|
import { describe, test, expect, beforeAll, afterAll } from 'bun:test';
|
|
import { CAPTURE_MS, CAPTURE_LONG_MS } from './helpers/eval-budgets';
|
|
import { runSkillTest, type SkillTestResult } from './helpers/session-runner';
|
|
import {
|
|
ROOT, runId,
|
|
describeIfSelected, testConcurrentIfSelected,
|
|
copyDirSync, logCost, recordE2E,
|
|
createEvalCollector, finalizeEvalCollector,
|
|
} from './helpers/e2e-helpers';
|
|
import { e2eTierEnabled } from './helpers/e2e-gate';
|
|
import { asideAvailable } from './helpers/aside-available';
|
|
import { startTestServer } from '../browse/test/test-server';
|
|
import * as fs from 'fs';
|
|
import * as path from 'path';
|
|
import * as os from 'os';
|
|
|
|
// --- Whole-file gate: periodic tier AND a live Aside. Skip, never fail. ---
|
|
|
|
const tierOk = e2eTierEnabled('periodic');
|
|
const shouldRun = tierOk && asideAvailable();
|
|
if (process.env.EVALS && !shouldRun) {
|
|
process.stderr.write(`\nskill-e2e-aside: SKIPPED — ${tierOk
|
|
? 'Aside is not installed or not running (or GSTACK_SKIP_ASIDE=1)'
|
|
: 'external-service test, periodic tier only'}\n`);
|
|
}
|
|
|
|
const SUITE = 'Aside-driven skills E2E';
|
|
const evalCollector = shouldRun ? createEvalCollector('e2e-aside') : null;
|
|
|
|
/** describeIfSelected, forced to describe.skip when the whole-file gate is closed. */
|
|
const describeAside = (name: string, keys: string[], fn: () => void) =>
|
|
describeIfSelected(name, keys, fn, shouldRun ? undefined : []);
|
|
|
|
// --- Helpers ---
|
|
|
|
/** Build output and sources never reach the agent — only SKILL.md + runtime assets. */
|
|
const SKIP_DIRS = new Set(['dist', 'src', 'test', 'bin', 'scripts', 'node_modules']);
|
|
/** Template sources and design notes are not runtime assets either. */
|
|
const skipFile = (name: string) => name.endsWith('.tmpl') || /^PLAN-.*\.md$/.test(name);
|
|
/** Generated docs: the tree, or a `gen:skill-docs --out-dir` render (mirrors the skill tree, outputs only). */
|
|
const DOCS_ROOT = process.env.GSTACK_E2E_DOCS_ROOT || ROOT;
|
|
|
|
function copySkill(name: string, dir: string): void {
|
|
const src = path.join(ROOT, name);
|
|
const dest = path.join(dir, name);
|
|
fs.mkdirSync(dest, { recursive: true });
|
|
for (const e of fs.readdirSync(src, { withFileTypes: true })) {
|
|
if (SKIP_DIRS.has(e.name) || skipFile(e.name)) continue;
|
|
if (e.isDirectory()) copyDirSync(path.join(src, e.name), path.join(dest, e.name));
|
|
else fs.copyFileSync(path.join(src, e.name), path.join(dest, e.name));
|
|
}
|
|
if (DOCS_ROOT !== ROOT && fs.existsSync(path.join(DOCS_ROOT, name))) copyDirSync(path.join(DOCS_ROOT, name), dest);
|
|
const doc = fs.readFileSync(path.join(dest, 'SKILL.md'), 'utf-8');
|
|
// A doc with the Aside contract AND the `$B` fallback is the shipped shape.
|
|
const stale = !doc.includes('BROWSER SETUP (Aside');
|
|
if (stale) throw new Error(`${name}/SKILL.md is not Aside-native (stale generated doc) — run \`bun run gen:skill-docs\` (or point GSTACK_E2E_DOCS_ROOT at an --out-dir render) before this E2E`);
|
|
}
|
|
|
|
/**
|
|
* Concatenated text of every BASH tool_result in the stream-json transcript.
|
|
* runSkillTest leaves toolCalls[].output empty; the agent's Bash INPUT and the
|
|
* skill doc it Reads both contain the sentinel string — only a Bash RESULT
|
|
* proves an `aside repl` script printed it.
|
|
*/
|
|
function bashOutput(result: SkillTestResult): string {
|
|
const bashIds = new Set<string>();
|
|
const parts: string[] = [];
|
|
for (const e of result.transcript) {
|
|
for (const item of e?.message?.content ?? []) {
|
|
if (e.type === 'assistant' && item?.type === 'tool_use' && item.name === 'Bash') bashIds.add(item.id);
|
|
if (e.type === 'user' && item?.type === 'tool_result' && bashIds.has(item.tool_use_id)) {
|
|
const c = item.content;
|
|
parts.push(typeof c === 'string' ? c : (Array.isArray(c) ? c.map((x: any) => x?.text ?? '').join('\n') : ''));
|
|
}
|
|
}
|
|
}
|
|
return parts.join('\n');
|
|
}
|
|
|
|
// Line-anchored: a `cat SKILL.md` in Bash prints `console.log("GSTACK_STEP_OK");`, never a bare token line.
|
|
const STEP_OK = /^GSTACK_STEP_OK\s*$/m;
|
|
const DIFF_START = /^DIFF_START\s*$/m;
|
|
const CONSOLE_ERRORS = /^CONSOLE_ERRORS=\[/m;
|
|
|
|
/** Every case: at least one `aside repl` script ran to completion. */
|
|
function expectDroveAside(result: SkillTestResult): string {
|
|
const out = bashOutput(result);
|
|
expect(out).toMatch(STEP_OK);
|
|
return out;
|
|
}
|
|
|
|
/** First parseable JSON document in the final message: whole text, a fenced block, or the outermost [...] / {...}. */
|
|
function extractJson(text: string): unknown {
|
|
const t = text.trim();
|
|
const candidates = [t, ...[...t.matchAll(/```(?:json)?\s*([\s\S]*?)```/g)].map((m) => m[1].trim())];
|
|
for (const [open, close] of [['[', ']'], ['{', '}']]) {
|
|
const a = t.indexOf(open);
|
|
const b = t.lastIndexOf(close);
|
|
if (a !== -1 && b > a) candidates.push(t.slice(a, b + 1));
|
|
}
|
|
for (const c of candidates) {
|
|
try { return JSON.parse(c); } catch { /* try the next shape */ }
|
|
}
|
|
return null;
|
|
}
|
|
|
|
function firstArray(doc: unknown): unknown[] | null {
|
|
if (Array.isArray(doc)) return doc;
|
|
if (doc && typeof doc === 'object') {
|
|
for (const v of Object.values(doc as Record<string, unknown>)) if (Array.isArray(v)) return v;
|
|
}
|
|
return null;
|
|
}
|
|
|
|
const filesIn = (dir: string): string[] => (fs.existsSync(dir) ? fs.readdirSync(dir) : []);
|
|
|
|
/** Common preface: the fixture server is up, Aside is up, drive it through the skill's own cookbook. */
|
|
const preface = (skillMd: string) => `Aside is installed and running on this machine; its BROWSER SETUP probe will print READY.
|
|
Read ${skillMd} and follow its BROWSER SETUP and cookbook exactly. Skip the preamble bash block, lake intro, telemetry, and contributor-mode sections — go straight to the workflow.
|
|
Drive the browser ONLY with 'aside repl' scripts, each ending with console.log("GSTACK_STEP_OK"); copy any screenshot out of the printed ASIDE_DIR in bash.
|
|
Do not use AskUserQuestion. The target is a local test server that is already running — do not start servers or discover ports.
|
|
`;
|
|
|
|
// --- Cases ---
|
|
|
|
interface AsideCase {
|
|
key: string;
|
|
skills: string[];
|
|
maxTurns: number;
|
|
timeout: number;
|
|
prompt: (dir: string, url: string) => string;
|
|
check: (dir: string, result: SkillTestResult) => void;
|
|
}
|
|
|
|
const CASES: AsideCase[] = [
|
|
{
|
|
key: 'aside-browse-basic',
|
|
skills: ['browse'],
|
|
maxTurns: 30,
|
|
timeout: CAPTURE_MS,
|
|
prompt: (_dir, url) => `${preface('browse/SKILL.md')}
|
|
Open ${url}/basic.html, report the page title, the interactive elements from the snapshot tree, and the console error count, then close the tab.
|
|
Your final message must quote the page title verbatim.`,
|
|
check: (_dir, result) => {
|
|
expect(result.exitReason).toBe('success');
|
|
expect(result.output.toLowerCase()).toContain('test page - basic');
|
|
expect(expectDroveAside(result)).toMatch(CONSOLE_ERRORS);
|
|
},
|
|
},
|
|
{
|
|
key: 'aside-browse-flow',
|
|
skills: ['browse'],
|
|
maxTurns: 30,
|
|
timeout: CAPTURE_MS,
|
|
prompt: (dir, url) => `${preface('browse/SKILL.md')}
|
|
Drive this flow in ONE script, shaped like the cookbook's "Drive a flow" recipe:
|
|
open ${url}/forms.html, take a baseline interactive snapshot, fill #name with "Aside QA", select "user" in #role, click #profile-btn, wait for #result to be visible, then print the DIFF_START/DIFF_END block, the URL= line and the CONSOLE_ERRORS= line, save a screenshot named "flow-result.jpg", print ASIDE_DIR=, close the tab, print the sentinel.
|
|
Then, in bash, mkdir -p ${dir}/screenshots and copy flow-result.jpg from the printed ASIDE_DIR into ${dir}/screenshots/.
|
|
Your final message must include the DIFF and CONSOLE_ERRORS lines verbatim.`,
|
|
check: (dir, result) => {
|
|
expect(result.exitReason).toBe('success');
|
|
const out = expectDroveAside(result);
|
|
expect(out).toMatch(DIFF_START);
|
|
expect(out).toMatch(CONSOLE_ERRORS);
|
|
const shots = filesIn(path.join(dir, 'screenshots')).filter((f) => /\.(jpe?g|png)$/i.test(f));
|
|
expect(shots.length).toBeGreaterThan(0);
|
|
expect(fs.statSync(path.join(dir, 'screenshots', shots[0])).size).toBeGreaterThan(0);
|
|
},
|
|
},
|
|
{
|
|
key: 'aside-qa-quick',
|
|
skills: ['qa'],
|
|
maxTurns: 40,
|
|
timeout: CAPTURE_LONG_MS,
|
|
prompt: (dir, url) => `${preface('qa/SKILL.md')}
|
|
qa is a carved skill: when SKILL.md tells you to Read ~/.claude/skills/gstack/qa/sections/<file>, read qa/sections/<file> in this working directory instead (same content, local copy).
|
|
Also skip the clean-working-tree check and the test-framework bootstrap — this directory has no source code and nothing to fix, so run a report-only pass.
|
|
Run a Quick-tier QA test on ${url}/basic.html.
|
|
Output dir: ${dir}/qa-reports — write the report to ${dir}/qa-reports/qa-report.md (it must include the Health Score line) and copy every screenshot into ${dir}/qa-reports/screenshots/.`,
|
|
check: (dir, result) => {
|
|
// Thorough QA may run out of turns; the artifacts are the contract.
|
|
expect(['success', 'error_max_turns']).toContain(result.exitReason);
|
|
expectDroveAside(result);
|
|
const report = path.join(dir, 'qa-reports', 'qa-report.md');
|
|
expect(fs.existsSync(report)).toBe(true);
|
|
expect(fs.readFileSync(report, 'utf-8')).toMatch(/health score/i);
|
|
expect(filesIn(path.join(dir, 'qa-reports', 'screenshots')).length).toBeGreaterThan(0);
|
|
},
|
|
},
|
|
{
|
|
key: 'aside-scrape-json',
|
|
skills: ['scrape'],
|
|
maxTurns: 30,
|
|
timeout: CAPTURE_MS,
|
|
prompt: (_dir, url) => `${preface('scrape/SKILL.md')}
|
|
Scrape the list of links (text and href) on ${url}/basic.html.
|
|
Follow the skill's output discipline: your final message is exactly one JSON document — an object with an "items" array (or a bare array) — with no prose around it.`,
|
|
check: (_dir, result) => {
|
|
expect(result.exitReason).toBe('success');
|
|
expectDroveAside(result);
|
|
const doc = extractJson(result.output);
|
|
expect(doc).not.toBeNull();
|
|
const items = firstArray(doc);
|
|
expect(items).not.toBeNull();
|
|
expect(items!.length).toBeGreaterThan(0);
|
|
expect(JSON.stringify(doc)).toContain('page1');
|
|
},
|
|
},
|
|
{
|
|
key: 'aside-canary-quick',
|
|
skills: ['canary'],
|
|
maxTurns: 40,
|
|
timeout: CAPTURE_LONG_MS,
|
|
prompt: (dir, url) => `${preface('canary/SKILL.md')}
|
|
Run: /canary ${url}/basic.html --quick — a single-pass health check on that one page, no continuous monitoring, no baseline capture.
|
|
The working directory is ${dir}; write the report to ${dir}/.gstack/canary-reports/<date>-canary.md (and the .json) exactly as the skill describes. If a gstack helper binary is missing, skip that logging line and continue.`,
|
|
check: (dir, result) => {
|
|
expect(['success', 'error_max_turns']).toContain(result.exitReason);
|
|
expectDroveAside(result);
|
|
const reportDir = path.join(dir, '.gstack', 'canary-reports');
|
|
const reports = filesIn(reportDir).filter((f) => /-canary\.md$/.test(f));
|
|
expect(reports.length).toBeGreaterThan(0);
|
|
expect(fs.readFileSync(path.join(reportDir, reports[0]), 'utf-8')).toMatch(/CANARY REPORT|Status:/);
|
|
},
|
|
},
|
|
];
|
|
|
|
// --- Suite: one describe so the sessions run concurrently (one shard wall for five captures) ---
|
|
|
|
let server: ReturnType<typeof startTestServer>;
|
|
|
|
describeAside(SUITE, CASES.map((c) => c.key), () => {
|
|
beforeAll(() => { server = startTestServer(); });
|
|
afterAll(() => { server?.server?.stop(); });
|
|
|
|
for (const c of CASES) {
|
|
testConcurrentIfSelected(c.key, async () => {
|
|
const dir = fs.mkdtempSync(path.join(os.tmpdir(), `skill-e2e-${c.key}-`));
|
|
try {
|
|
for (const s of c.skills) copySkill(s, dir);
|
|
if (c.skills.includes('qa')) fs.mkdirSync(path.join(dir, 'qa-reports', 'screenshots'), { recursive: true });
|
|
const result = await runSkillTest({
|
|
prompt: c.prompt(dir, server.url),
|
|
workingDirectory: dir,
|
|
maxTurns: c.maxTurns,
|
|
timeout: c.timeout,
|
|
testName: c.key,
|
|
runId,
|
|
});
|
|
logCost(`/${c.key}`, result);
|
|
let passed = false;
|
|
try {
|
|
c.check(dir, result);
|
|
passed = true;
|
|
} finally {
|
|
recordE2E(evalCollector, c.key, SUITE, result, { passed });
|
|
}
|
|
} finally {
|
|
fs.rmSync(dir, { recursive: true, force: true });
|
|
}
|
|
}, c.timeout);
|
|
}
|
|
});
|
|
|
|
// Explicit 60s timeout: finalize does a JSON save + cross-run comparison,
|
|
// observed past bun's 5s default hook timeout in sibling files.
|
|
afterAll(async () => {
|
|
await finalizeEvalCollector(evalCollector);
|
|
}, 60_000);
|