mirror of
https://github.com/stack-auth/stack.git
synced 2026-06-30 21:01:54 +08:00
0a23409a87
- Added a script for building a shared config-agent base snapshot, optimizing the warm-boot process for configuration updates. - Introduced a new script to link existing projects to GitHub repositories without re-seeding, improving workflow efficiency. - Updated the workflow paths and configuration file names to align with the new Hexclave structure. - Refactored existing scripts to ensure consistency in configuration paths and enhance overall integration with GitHub. Co-Authored-By: mantra <mantra@stack-auth.com>
74 lines
2.6 KiB
TypeScript
74 lines
2.6 KiB
TypeScript
/**
|
|
* End-to-end smoke test for the config-update repo agent against a REAL Vercel
|
|
* Sandbox and a REAL GitHub repo. This is a scratch script, not a unit test.
|
|
*
|
|
* It runs the single-phase apply flow: boot a sandbox (warm from the shared base
|
|
* snapshot if STACK_CONFIG_AGENT_BASE_SNAPSHOT_ID is set, else cold-install the
|
|
* agent SDK), clone the repo, agent edits the config, COMMIT + PUSH to the branch.
|
|
*
|
|
* WARNING: this pushes a commit to `main` of the target repo. Point it at a
|
|
* throwaway repo (default: hexclave/stackframe-website-2026).
|
|
*
|
|
* Run from apps/backend:
|
|
* pnpm dlx dotenv-cli -e .env.development.local -e .env.development -- \
|
|
* pnpm tsx scripts/spike-orchestrator-e2e.mts
|
|
*
|
|
* The GitHub token comes from $GITHUB_TOKEN, falling back to `gh auth token`.
|
|
*/
|
|
import { execFileSync } from "child_process";
|
|
import {
|
|
applyConfigUpdate,
|
|
type GithubRepoRef,
|
|
} from "../src/lib/config/repo-agent";
|
|
|
|
const REF: GithubRepoRef = {
|
|
owner: process.env.SPIKE_OWNER ?? "hexclave",
|
|
repo: process.env.SPIKE_REPO ?? "stackframe-website-2026",
|
|
branch: process.env.SPIKE_BRANCH ?? "main",
|
|
};
|
|
|
|
// The COMPLETE config we want the repo's config file to reflect (the real flow
|
|
// computes this from the branch config override; here it's a small literal).
|
|
const COMPLETE_CONFIG = { auth: { allowSignUp: false } } as Record<string, unknown>;
|
|
|
|
function githubToken(): string {
|
|
if (process.env.GITHUB_TOKEN) return process.env.GITHUB_TOKEN;
|
|
return execFileSync("gh", ["auth", "token"], { encoding: "utf-8" }).trim();
|
|
}
|
|
|
|
function ts(): string {
|
|
return new Date().toISOString().slice(11, 19);
|
|
}
|
|
function log(msg: string) {
|
|
console.log(`[${ts()}] ${msg}`);
|
|
}
|
|
|
|
async function main() {
|
|
// Fresh token per boot (matches the production provider seam). Here it's the
|
|
// same static token, fetched lazily each time the orchestrator boots a sandbox.
|
|
const getGithubToken = async () => githubToken();
|
|
log(`Target: ${REF.owner}/${REF.repo}@${REF.branch}`);
|
|
|
|
log("applyConfigUpdate (boot + clone + agent edit + push)…");
|
|
const t2 = Date.now();
|
|
const result = await applyConfigUpdate({
|
|
getGithubToken,
|
|
ref: REF,
|
|
completeConfig: COMPLETE_CONFIG,
|
|
commitMessage: "chore(hexclave): e2e smoke — set auth.allowSignUp=false",
|
|
});
|
|
log(`Done in ${((Date.now() - t2) / 1000).toFixed(0)}s`);
|
|
log(`Result: ${JSON.stringify(result)}`);
|
|
|
|
if (result.mode === "commit-to-branch") {
|
|
log(`✅ Pushed: ${result.commitUrl}`);
|
|
} else {
|
|
log("⚠️ Agent produced no change (config already matched).");
|
|
}
|
|
}
|
|
|
|
main().catch((err) => {
|
|
console.error("E2E FAILED:", err);
|
|
process.exit(1);
|
|
});
|