_check-deps
Skill50 starsChanged 21 days agoApache-2.0View on GitHub
Internal helper to check project dependencies (DB, email, etc.) with robust heuristics that don't fall for T3 bootstrap placeholders or localhost defaults. Delegates to bundled scripts/check-deps.mjs. Returns JSON. Triggered by add-auth, add-backup-db, and any skill that needs to verify a real cloud dependency is wired up. Not meant to be invoked directly by users.
- Reads credentials
1 other project uses a version of this.
Adopted by
Most recently taken up by
- flavien-ia/hypervibe-harnessword for word, first seen Oct 23, 2026
Other versions1 project uses a version of this.
Word for word · 1
- flavien-ia/hypervibe-harness★ 50100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository89 other files it gives its agents.
- accessibilitySkill
- add-2faSkill
- add-agent-dashboardSkill
- add-aiSkill
- add-analyticsSkill
- add-authSkill
- add-automationSkill
- add-backup-dbSkill
- add-collabSkill
- add-cronSkill
- add-dark-modeSkill
- add-dbSkill
- add-domainSkill
- add-emailSkill
- add-github-authSkill
- add-google-authSkill
- add-i18nSkill
- _add-keyringSkill
- add-mapSkill
- add-notification-centerSkill
- add-push-notificationSkill
- add-pwaSkill
- add-roleSkill
- add-routineSkill
- add-storageSkill
- add-stripeSkill
- add-testSkill
- bootstrapSkill
- cleanSkill
- _collect-secretSkill
- _convert-to-turborepoSkill
- _create-agentSkill
- _create-cloudflare-workerSkill
- _create-contact-pageSkill
- _create-render-workerSkill
- _create-routineSkill
- _create-workflowSkill
- delete-projectSkill
- _detect-project-rootSkill
- _dns-brevoSkill
- _dns-cloudflareSkill
- _dns-gandiSkill
- _dns-godaddy-manualSkill
- _dns-hostingerSkill
- _dns-infomaniak-manualSkill
- _dns-ionos-manualSkill
- _dns-lwsSkill
- _dns-namecheapSkill
- _dns-ovhSkill
- _dns-porkbunSkill
- _dns-resendSkill
- _dns-squarespace-manualSkill
- eco-auditSkill
- _ensure-aiSkill
- _ensure-vaultSkill
- _generate-secretSkill
- geoSkill
- _get-secretSkill
- gscSkill
- _hash-passwordSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Use plain, non-technical business language.
- Distinctive: line 14
When generating user-facing content for the scaffolded project (UI labels, emails, copy), write it in the user's language too.
- Distinctive: line 17
Check if a project has real cloud dependencies wired up (not just T3 defaults / placeholders).
- Very specific: line 48
If --include-vercel is passed, the JSON output includes a _meta key with :
- Very specific: line 139
If the check fails, the reason includes a suggestion : - run /start to configure.
- Very specific: line 155
Always: use this helper when a skill's next step depends on a dependency being real (never grep DATABASE_URL .env inline - T3 placeholders …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Check Deps - Internal helper
Communication
- Detect the user's language from the conversation (the user's own messages, anywhere in the session - not just this invocation: a bare slash command like
/bootstrapcarries no language signal by itself). If nothing in the conversation gives a signal, fall back to the OS locale (node -e "console.log(Intl.DateTimeFormat().resolvedOptions().locale)") before defaulting to English. ALWAYS reply in that language for every user-facing message: questions, progress, confirmations, summaries, errors - including any example text quoted in this skill, which is illustrative and must be translated, never sent verbatim. - Use plain, non-technical business language. Never expose internal script names (*.mjs) or jargon; describe actions in human terms.
- When generating user-facing content for the scaffolded project (UI labels, emails, copy), write it in the user's language too.
- Show progress as a short natural-language checklist (in-progress and done states).
Check if a project has real cloud dependencies wired up (not just T3 defaults / placeholders). Delegates to a bundled Node script - never reimplement the checks inline.
Env files read
The script merges all Next.js-style env files found at the cwd, with Next.js precedence (later overrides earlier) :
.env.env.development.env.development.local.env.local
So a var set in any of these is detected - critical since some projects put DATABASE_URL in .env.local only.
Invocation
From the project root :
node "${CLAUDE_SKILL_DIR}/../../scripts/check-deps.mjs" <check1> [<check2> ...] [--include-vercel]Output : JSON on stdout. Exit code is always 0 - parse the JSON to get per-check ok + reason.
Flag --include-vercel
Opt-in : runs a vercel env pull on the production environment and merges the Vercel vars on top of the local vars (Vercel wins on conflict). Useful for checks like "is production correctly configured?" (e.g., /security, /clean).
Prerequisites : project linked to Vercel (.vercel/project.json present) and Vercel CLI installed.
Cost : ~2 seconds of network call. Only use it when the Vercel source matters.
If --include-vercel is passed, the JSON output includes a _meta key with :
sources: ["local", "vercel-production"]if the pull succeeded, otherwise["local"]onlyvercelPull.keys: list of the names of the vars retrieved from Vercel (never the values - no secret leak in the output)
Whatever the checks requested, _meta.sharedWorker is present when a shared clock was deployed from this machine. stale: true means it runs an older plugin version, which happens after a manual install of the plugin: say it once, in one sentence, with the remedy in hint. Never redeploy from here.
Supported checks
db - real cloud DB wired up
ok: true if and only if :
DATABASE_URLis present in.env- Its value matches NONE of the disqualifying patterns (case-insensitive) :
@localhost:(points to local)@127.0.0.1:(same)placeholder(the word, wherever it is)//postgres:postgres@(typical T3/Docker default pair)YOUR_DB(marker from.env.example)^file:(local SQLite, not a cloud DB)
- A
drizzle.config.tsor.jsexists at the root, inapps/web/, or inpackages/db/
Fields returned : { ok, reason, host?, drizzleConfig? }.
email - Resend or Brevo configured
ok: true if RESEND_API_KEY OR BREVO_API_KEY is present in .env AND does not match a placeholder pattern (empty, placeholder, your_api_key, xxx..., re_your..., xkeysib-your...).
Fields returned : { ok, provider: "resend"|"brevo"|null, reason }.
auth - NextAuth / auth lib installed & configured
ok: true if :
- An auth file exists at one of these locations (at the root level OR prefixed by
apps/web/) :src/server/auth.ts,src/server/auth/index.ts,src/server/auth.config.ts,src/lib/auth.ts,src/lib/auth/index.ts,src/auth.ts,src/auth/index.ts,src/app/auth.ts,auth.ts,auth.config.ts - A secret is present in the env :
AUTH_SECRETORNEXTAUTH_SECRETORBETTER_AUTH_SECRET, and non-placeholder
The check also tries to infer the mode from the env : presence of ADMIN_PASSWORD_HASH_DEV or _PROD → admin-credentials, otherwise user-credentials.
Fields returned : { ok, reason, authFile?, secretVar?, mode?: "admin-credentials"|"user-credentials" }.
vercel - project linked to Vercel
ok: true if .vercel/project.json exists at the root and contains valid projectId + orgId.
Fields returned : { ok, reason, projectId?, orgId? }.
github-repo - project pushed to a GitHub remote
ok: true if .git/config contains a remote pointing to github.com (HTTPS or SSH). Parses the owner and the repo name.
Fields returned : { ok, reason, owner?, repo?, nameWithOwner? }.
i18n - next-intl installed
ok: true if next-intl is listed in the dependencies of package.json (root) or apps/web/package.json. Also tries to locate the messages/ folder for the translations.
Fields returned : { ok, reason, packageJson?, messagesDir?: string|null }.
stripe - Stripe configured
ok: true if STRIPE_SECRET_KEY is present, non-placeholder, and starts with sk_. Automatically detects the mode (test or live).
Fields returned : { ok, reason, mode?: "test"|"live"|"unknown" }.
storage - Cloudflare R2 configured
ok: true if R2_ACCOUNT_ID (or CLOUDFLARE_ACCOUNT_ID), R2_ACCESS_KEY_ID, and R2_SECRET_ACCESS_KEY are all present and non-placeholders.
Also detects the R2 jurisdiction from the format of R2_ENDPOINT :
https://<acc>.eu.r2.cloudflarestorage.com→jurisdiction: "eu"(strict GDPR ✅)https://<acc>.r2.cloudflarestorage.com→jurisdiction: "default"(global jurisdiction, data potentially outside the EU)- absent →
jurisdiction: "unknown"
If the jurisdiction is not EU, also returns a jurisdictionWarning field (string) with a migration recommendation. ok stays true so as not to break existing setups - it is up to the consuming skill to surface the warning if relevant (e.g., in the Step 0 menu of add-storage).
Fields returned : { ok, reason, bucket?: string|null, publicUrl?: string|null, jurisdiction?: "eu"|"default"|"unknown", jurisdictionWarning?: string|null }.
analytics - Google Analytics (GA4) configured
ok: true if NEXT_PUBLIC_GA_MEASUREMENT_ID (the name /add-analytics writes; the older NEXT_PUBLIC_GA_ID is still accepted) is present, non-placeholder, and starts with G- (GA4 format).
Fields returned : { ok, reason, gaId? }.
cloudflare - Cloudflare API token configured AND valid
Checks that a Cloudflare API token is in place in the system env (not in .env, because it is a machine config shared across projects) AND that it is actually valid by querying https://api.cloudflare.com/client/v4/user/tokens/verify.
ok: true if and only if :
CLOUDFLARE_API_TOKENORCF_API_TOKENis present (hypervibe convention = both point to the same token -wranglerreadsCLOUDFLARE_API_TOKEN, thecurlREST API calls can use either name)- The value is not a placeholder
- The Cloudflare API returns
{ success: true, result: { status: "active" } }for this token
If the check fails, the reason includes a suggestion : - run /start to configure. To be used before any Cloudflare operation (DNS via curl API, Workers via wrangler, Email Routing via curl, R2 via wrangler or curl...).
Fields returned : { ok, reason, varName? }.
Typical usage
result=$(node "${CLAUDE_SKILL_DIR}/../../scripts/check-deps.mjs" db email)
# Parse with node -e (jq is not installed on the target machine)
db_ok=$(echo "$result" | node -e "console.log(JSON.parse(require('fs').readFileSync(0,'utf8')).db.ok)")
email_ok=$(echo "$result" | node -e "console.log(JSON.parse(require('fs').readFileSync(0,'utf8')).email.ok)")
email_provider=$(echo "$result" | node -e "console.log(JSON.parse(require('fs').readFileSync(0,'utf8')).email.provider || 'none')")Rules
- Always use this helper when a skill's next step depends on a dependency being real (never
grep DATABASE_URL .envinline - T3 placeholders systematically cause false positives). - If a check returns
ok: false, relay thereasonto the user in plain language (translate from technical to non-tech - ex: "DATABASE_URL points to localhost" → "your database points to your own computer, you need one reachable from the internet"). Then offer to invoke the correspondingadd-*skill via a natural-language prompt. - Exit code is always 0 - the script never fails just because a check is negative. Only a truly malformed invocation (unknown flag) exits non-zero.
Extending
To add a new check (e.g., auth, vercel-linked, github-repo), add a function in scripts/check-deps.mjs and wire it into the dispatch switch. Document the new check here in the "Supported checks" section. Keep heuristics encapsulated - never require callers to reimplement them.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

