GraphQL Clinic — API

Paste the SDL and resolvers, get a senior API review and a full refined schema.

API tokens Open the app

Review your GraphQL schema and resolvers from your own scripts

Send GraphQL SDL and resolver code — a type, a schema file, a resolver map, or a whole grab-bag — and get back one JSON object: an honest sound / refactor / rework verdict, a health check across five API areas, findings ranked by severity each with corrected GraphQL, a twelve-item checklist scored against the paste, and a complete refined rewrite of what you pasted. Everything this app does goes through the SkillSafe App API — plain JSON over HTTPS — so you can wire the review into a CI gate, a pull-request bot, or a pre-merge check that refuses a schema whose list field is an unpaginated posts: [Post]. Every code step below is shown in cURL, Python, JavaScript, Go, Java, Ruby, PHP and C#; pick a language once and the whole page follows.

Basics

Base URL: https://api.skillsafe.ai/v1/app-api, app slug graphql-clinic. Every request sends Authorization: Bearer <token> and JSON bodies with Content-Type: application/json. Responses are wrapped in an envelope: {"data": …} on success, {"error": {"code", "message"}} on failure. The review itself is produced by the gpt-terra model. Estimates are free; runs are metered against your credit balance. There is a single run task — one paste in, one review out, no follow-up calls and no session state to carry.

StatusMeaning
401Missing or expired token — create a new session.
402Not enough credits — top up at skillsafe.ai/account/credits.
403The token isn't allowed to do this (e.g. a guest reviewing a very large paste).
404Unknown job or record id.
5xxTransient platform error — retry with backoff.

Browsers enforce CORS for this API, so run these examples from a server, script or terminal — not from another website's frontend.

Step 0 — A tiny client

Every task below is a single HTTP call, so start with a short helper that adds the auth header, sends JSON and unwraps the data envelope. The later steps reuse it.

export API="https://api.skillsafe.ai/v1/app-api"
export TOKEN="YOUR_TOKEN"      # see step 1

# every call looks like:
#   curl -s "$API/..." -H "Authorization: Bearer $TOKEN" [-d '{json}']
# jq is used below to pull fields out of the {"data": ...} envelope
import json, requests

API = "https://api.skillsafe.ai/v1/app-api"
TOKEN = "YOUR_TOKEN"  # see step 1 — read it from your shell environment in real code

def api(method, path, body=None, **headers):
    res = requests.request(method, API + path, json=body,
                           headers={"Authorization": f"Bearer {TOKEN}", **headers})
    payload = res.json()
    if not res.ok:
        raise RuntimeError(payload.get("error", {}).get("message", res.reason))
    return payload["data"]
// Node 18+ (built-in fetch)
const API = "https://api.skillsafe.ai/v1/app-api";
const TOKEN = "YOUR_TOKEN"; // see step 1 — read it from your shell environment in real code

async function api(method, path, body, extraHeaders = {}) {
  const res = await fetch(API + path, {
    method,
    headers: { Authorization: `Bearer ${TOKEN}`, "Content-Type": "application/json", ...extraHeaders },
    body: body === undefined ? undefined : JSON.stringify(body),
  });
  const json = await res.json();
  if (!res.ok) throw new Error(json.error?.message ?? res.statusText);
  return json.data;
}
package main

import (
	"bytes"
	"encoding/json"
	"fmt"
	"net/http"
	"os"
)

const API = "https://api.skillsafe.ai/v1/app-api"

var token = os.Getenv("SKILLSAFE_TOKEN") // see step 1

func call(method, path string, body, out any) error {
	var buf bytes.Buffer
	if body != nil {
		json.NewEncoder(&buf).Encode(body)
	}
	req, _ := http.NewRequest(method, API+path, &buf)
	req.Header.Set("Authorization", "Bearer "+token)
	req.Header.Set("Content-Type", "application/json")
	res, err := http.DefaultClient.Do(req)
	if err != nil {
		return err
	}
	defer res.Body.Close()
	var env struct {
		Data  json.RawMessage `json:"data"`
		Error *struct{ Message string `json:"message"` } `json:"error"`
	}
	json.NewDecoder(res.Body).Decode(&env)
	if res.StatusCode >= 400 {
		return fmt.Errorf("api %s %s: %s", method, path, env.Error.Message)
	}
	if out == nil {
		return nil
	}
	return json.Unmarshal(env.Data, out)
}
// Java 17+, no dependencies. Pair with your JSON library (Jackson, Gson…)
// to read fields out of the returned envelope.
import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;

public class SkillSafe {
    static final String API = "https://api.skillsafe.ai/v1/app-api";
    static final String TOKEN = System.getenv("SKILLSAFE_TOKEN"); // see step 1
    static final HttpClient HTTP = HttpClient.newHttpClient();

    static String api(String method, String path, String jsonBody) throws Exception {
        var req = HttpRequest.newBuilder(URI.create(API + path))
            .header("Authorization", "Bearer " + TOKEN)
            .header("Content-Type", "application/json")
            .method(method, jsonBody == null
                ? HttpRequest.BodyPublishers.noBody()
                : HttpRequest.BodyPublishers.ofString(jsonBody))
            .build();
        var res = HTTP.send(req, HttpResponse.BodyHandlers.ofString());
        if (res.statusCode() >= 400) throw new RuntimeException(res.body());
        return res.body(); // envelope: {"data": …}
    }
}
require "net/http"
require "json"

API = "https://api.skillsafe.ai/v1/app-api"
TOKEN = ENV.fetch("SKILLSAFE_TOKEN") # see step 1

def api(method, path, body = nil)
  uri = URI(API + path)
  req = Net::HTTP.const_get(method.capitalize).new(uri)
  req["Authorization"] = "Bearer #{TOKEN}"
  req["Content-Type"] = "application/json"
  req.body = body.to_json if body
  res = Net::HTTP.start(uri.host, uri.port, use_ssl: true) { |h| h.request(req) }
  payload = JSON.parse(res.body)
  raise (payload.dig("error", "message") || res.message) unless res.is_a?(Net::HTTPSuccess)
  payload["data"]
end
<?php
const API = "https://api.skillsafe.ai/v1/app-api";
$TOKEN = getenv("SKILLSAFE_TOKEN"); // see step 1

function api(string $method, string $path, ?array $body = null): mixed {
    global $TOKEN;
    $ch = curl_init(API . $path);
    curl_setopt_array($ch, [
        CURLOPT_CUSTOMREQUEST  => $method,
        CURLOPT_RETURNTRANSFER => true,
        CURLOPT_HTTPHEADER     => [
            "Authorization: Bearer $TOKEN",
            "Content-Type: application/json",
        ],
        CURLOPT_POSTFIELDS     => $body === null ? null : json_encode($body),
    ]);
    $payload = json_decode(curl_exec($ch), true);
    $status  = curl_getinfo($ch, CURLINFO_RESPONSE_CODE);
    curl_close($ch);
    if ($status >= 400) {
        throw new Exception($payload["error"]["message"] ?? "HTTP $status");
    }
    return $payload["data"];
}
// .NET 8+
using System.Net.Http.Json;
using System.Text.Json;

static class SkillSafe
{
    const string Api = "https://api.skillsafe.ai/v1/app-api";
    static readonly HttpClient Http = new();

    static SkillSafe() =>
        Http.DefaultRequestHeaders.Authorization =
            new("Bearer", Environment.GetEnvironmentVariable("SKILLSAFE_TOKEN")); // see step 1

    public static async Task<JsonElement> ApiAsync(HttpMethod method, string path, object? body = null)
    {
        var req = new HttpRequestMessage(method, Api + path);
        if (body != null) req.Content = JsonContent.Create(body);
        var res = await Http.SendAsync(req);
        var json = await res.Content.ReadFromJsonAsync<JsonElement>();
        if (!res.IsSuccessStatusCode)
            throw new Exception(json.GetProperty("error").GetProperty("message").GetString());
        return json.GetProperty("data");
    }
}

Step 1 — Get a token

POST /guest

A guest token lets you check balances and estimate costs for free. For metered review runs billed to your own account, use your personal token: open the token page, sign in with SkillSafe, and press Copy shell export — it puts export SKILLSAFE_TOKEN="…" on your clipboard, which every example below reads. Treat the token like a password: it can spend your credits. For fully headless scripts, POST /guest mints a guest token with no browser involved.

curl -s -X POST "$API/guest" \
  -H "Content-Type: application/json" \
  -d '{"slug":"graphql-clinic"}' | jq -r '.data.token'
token = api("POST", "/guest", {"slug": "graphql-clinic"})["token"]
const { token } = await api("POST", "/guest", { slug: "graphql-clinic" });
var guest struct{ Token string `json:"token"` }
err := call("POST", "/guest", map[string]string{"slug": "graphql-clinic"}, &guest)
String envelope = api("POST", "/guest", """
    {"slug":"graphql-clinic"}""");
// token is at data.token in the returned JSON
token = api("POST", "/guest", { slug: "graphql-clinic" })["token"]
$token = api("POST", "/guest", ["slug" => "graphql-clinic"])["token"];
var guest = await SkillSafe.ApiAsync(HttpMethod.Post, "/guest",
    new { slug = "graphql-clinic" });
var token = guest.GetProperty("token").GetString();

The app stores this browser's token under the localStorage key skillsafe_app_token:graphql-clinic, on the app's own origin. The token page reads and manages it for you — you never need to open developer tools.

Step 2 — Check who you are and your balance

GET /me

Returns subject_type ("user" or "guest"), subject_id and your credits balance. Check this before reviewing a large paste.

curl -s "$API/me" -H "Authorization: Bearer $TOKEN" | jq '.data'
me = api("GET", "/me")
print(me["subject_type"], me["credits"])
const me = await api("GET", "/me");
console.log(me.subject_type, me.credits);
var me struct {
	SubjectType string `json:"subject_type"`
	Credits     int64  `json:"credits"`
}
err := call("GET", "/me", nil, &me)
String envelope = api("GET", "/me", null);
// data.subject_type, data.credits
me = api("GET", "/me")
puts "#{me["subject_type"]}: #{me["credits"]} credits"
$me = api("GET", "/me");
echo "{$me['subject_type']}: {$me['credits']} credits\n";
var me = await SkillSafe.ApiAsync(HttpMethod.Get, "/me");
Console.WriteLine($"{me.GetProperty("subject_type")}: {me.GetProperty("credits")} credits");

Step 3 — Estimate the cost

POST /estimate

Send exactly the input you would send to /run; the response's hold_credits is the worst-case cost. Nothing is charged and no job is created, so estimating is free — useful when you are feeding in a whole diff or a directory of schema files and want a ceiling before spending credits.

Input fieldTypeNotes
codestring, requiredThe GraphQL SDL and/or resolver source to review, up to 100000 characters: a type, a schema file, a resolver map, or several files concatenated with file-name comment headers. Very long pastes may be clipped middle-out, with a [... clipped ...] marker showing where.
targetstringpublic | internal | federated — the API surface under review. The review is calibrated to it: public makes abuse limits, depth and complexity caps, gated introspection and opaque ids first-class concerns; internal leads with schema clarity, resolver efficiency and mutation contracts; federated adds subgraph boundaries, entity keys and cross-subgraph resolution.
notesstring, optionalExtra context, up to 20000 characters: what the API serves, the modelling intent, which parts are new, which clients you must not break.
prescan_factsobject, optionalWhat a client-side prescan mechanically detected in the code: {"antipatterns": [], "types": [], "signals": {}}. antipatterns and types hold {id, label, lines} entries — keyword-matched schema smells (ap:unpaginated-list, ap:bare-string-id) and type declarations found (t:User), each with the line numbers it was seen on. signals is a counter object: {"types": 0, "queryFields": 0, "mutationFields": 0, "listFields": 0, "inputTypes": 0, "enums": 0, "dataloaderUses": 0, "lines": 0}. Every id you send comes back in coverage_check. The web UI fills this from its own scan; API callers may omit the field or send {"antipatterns": [], "types": [], "signals": {}}.
retry_notestring, optionalOnly set by the app's automatic reformat retry when a first reply was not valid JSON. Leave it out.
cat > schema.graphql <<'GRAPHQL'
type Query { user(id: ID!): User }

type User {
  id: ID!
  posts: [Post]
}
GRAPHQL

jq -n --rawfile c schema.graphql \
  '{code: $c, target: "public", notes: "",
    prescan_facts: {antipatterns: [], types: [], signals: {}}}' > input.json

curl -s -X POST "$API/estimate" \
  -H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \
  -d @input.json | jq '.data.hold_credits'
CODE = """type Query { user(id: ID!): User }

type User {
  id: ID!
  posts: [Post]
}"""

payload = {
    "code": CODE,
    "target": "public",
    "notes": "",
    "prescan_facts": {"antipatterns": [], "types": [], "signals": {}},
}

est = api("POST", "/estimate", payload)
print("worst case:", est.get("hold_credits", est.get("credits")), "credits")
const code = `type Query { user(id: ID!): User }

type User {
  id: ID!
  posts: [Post]
}`;

const payload = {
  code,
  target: "public",
  notes: "",
  prescan_facts: { antipatterns: [], types: [], signals: {} },
};

const est = await api("POST", "/estimate", payload);
console.log("worst case:", est.hold_credits ?? est.credits, "credits");
const code = `type Query { user(id: ID!): User }

type User {
  id: ID!
  posts: [Post]
}`

payload := map[string]any{
	"code":   code,
	"target": "public",
	"notes":  "",
	"prescan_facts": map[string]any{
		"antipatterns": []any{}, "types": []any{}, "signals": map[string]any{},
	},
}

var est struct{ HoldCredits int64 `json:"hold_credits"` }
err := call("POST", "/estimate", payload, &est)
String code = """
    type Query { user(id: ID!): User }

    type User {
      id: ID!
      posts: [Post]
    }""";

String jsonPayload = """
    {"code": %s, "target": "public",
     "notes": "",
     "prescan_facts": {"antipatterns": [], "types": [], "signals": {}}}
    """.formatted(toJsonString(code));

String envelope = api("POST", "/estimate", jsonPayload);
// worst-case cost is at data.hold_credits
CODE_TEXT = <<~'GRAPHQL'
  type Query { user(id: ID!): User }

  type User {
    id: ID!
    posts: [Post]
  }
GRAPHQL

payload = { code: CODE_TEXT, target: "public",
            notes: "",
            prescan_facts: { antipatterns: [], types: [], signals: {} } }

est = api("POST", "/estimate", payload)
puts "worst case: #{est["hold_credits"] || est["credits"]} credits"
$code = <<<'GRAPHQL'
type Query { user(id: ID!): User }

type User {
  id: ID!
  posts: [Post]
}
GRAPHQL;

$payload = [
    "code"          => $code,
    "target"        => "public",
    "notes"         => "",
    "prescan_facts" => ["antipatterns" => [], "types" => [], "signals" => new stdClass()],
];

$est = api("POST", "/estimate", $payload);
echo "worst case: " . ($est["hold_credits"] ?? $est["credits"]) . " credits\n";
var code = """
    type Query { user(id: ID!): User }

    type User {
      id: ID!
      posts: [Post]
    }
    """;

var payload = new {
    code,
    target = "public",
    notes = "",
    prescan_facts = new {
        antipatterns = Array.Empty<object>(), types = Array.Empty<object>(),
        signals = new { },
    },
};

var est = await SkillSafe.ApiAsync(HttpMethod.Post, "/estimate", payload);
Console.WriteLine($"worst case: {est.GetProperty("hold_credits")} credits");

prescan_facts is how you make the review answer for things you already know about. Send {"antipatterns": [{"id": "ap:unpaginated-list", "label": "posts: [Post] returns an unbounded list", "lines": [5]}], "types": [{"id": "t:User", "label": "User (object type)", "lines": [3]}], "signals": {"types": 2, "queryFields": 1, "mutationFields": 0, "listFields": 1, "inputTypes": 0, "enums": 0, "dataloaderUses": 0, "lines": 5}} and every one of those ids comes back in coverage_check — addressed, or explained away as a false positive. Nothing you flag is silently dropped.

Step 4 — Run the review and wait for the result

POST /run
GET /jobs/{job_id}

/run takes the same input as /estimate, places a credit hold and returns a job_id. Poll /jobs/{job_id} every 1–2 seconds until status is succeeded or failed (a run typically takes 30–90 s, since the refined rewrite is written out in full). Always send an Idempotency-Key header so a network retry can't start a second, double-charged run. The review is in output — usually nested as output.output, and as a JSON string, so parse defensively. The samples below print the review name and verdict, the five health areas and the findings, then write rewrite.code to schema.refined.graphql using rewrite.filename.

JOB_ID=$(curl -s -X POST "$API/run" \
  -H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \
  -H "Idempotency-Key: review-$(date +%s)" \
  -d @input.json | jq -r '.data.job_id')

while :; do
  JOB=$(curl -s "$API/jobs/$JOB_ID" -H "Authorization: Bearer $TOKEN")
  STATUS=$(echo "$JOB" | jq -r '.data.status')
  [ "$STATUS" = "succeeded" ] || [ "$STATUS" = "failed" ] && break
  sleep 2
done

# unwrap the review once, then read it
echo "$JOB" | jq -r '.data.output.output' > review.json

jq -r '
  "\(.review_name) [\(.verdict_level)]: \(.verdict)",
  "",
  "HEALTH",
  (.health[] | "  [\(.status)] \(.area) - \(.note)"),
  "",
  "FINDINGS",
  (.findings[] | "  (\(.severity)) \(.category): \(.title)"),
  "",
  "CHECKLIST",
  (.checklist[] | "  [\(.status)] \(.item) - \(.note)")' review.json

# and drop the refined code straight into the repo
jq -r '.rewrite.code' review.json > "$(jq -r '.rewrite.filename' review.json)"   # schema.refined.graphql
import time

job_id = api("POST", "/run", payload,
             **{"Idempotency-Key": "review-001"})["job_id"]

while True:
    job = api("GET", f"/jobs/{job_id}")
    if job["status"] in ("succeeded", "failed"):
        break
    time.sleep(1.5)

if job["status"] == "failed":
    raise RuntimeError(job.get("error", "run failed"))

raw = job["output"]
if isinstance(raw, dict) and "output" in raw:
    raw = raw["output"]
review = json.loads(raw) if isinstance(raw, str) else raw

print(f'{review["review_name"]} [{review["verdict_level"]}]: {review["verdict"]}')
for area in review["health"]:
    print(f'  [{area["status"]:>4}] {area["area"]:<32} {area["note"]}')
for f in review["findings"]:
    print(f'  ({f["severity"]}) {f["category"]}: {f["title"]}')
    if f["fix_code"]:
        print(f'      {f["fix_code"]}')
for item in review["checklist"]:
    print(f'  [{item["status"]:>4}] {item["item"]:<42} {item["note"]}')
for c in review["coverage_check"]:
    print(f'  {c["id"]}: {"ok" if c["addressed"] else "SET ASIDE"} - {c["note"]}')

with open(review["rewrite"]["filename"], "w", encoding="utf-8") as fh:   # schema.refined.graphql
    fh.write(review["rewrite"]["code"])
import { writeFileSync } from "node:fs";

const { job_id } = await api("POST", "/run", payload,
  { "Idempotency-Key": crypto.randomUUID() });

let job;
do {
  await new Promise((r) => setTimeout(r, 1500));
  job = await api("GET", `/jobs/${job_id}`);
} while (job.status !== "succeeded" && job.status !== "failed");

if (job.status === "failed") throw new Error(job.error ?? "run failed");

const raw = job.output?.output ?? job.output;
const review = typeof raw === "string" ? JSON.parse(raw) : raw;

console.log(`${review.review_name} [${review.verdict_level}]: ${review.verdict}`);
for (const area of review.health) {
  console.log(`  [${area.status}] ${area.area}: ${area.note}`);
}
for (const f of review.findings) {
  console.log(`  (${f.severity}) ${f.category}: ${f.title}`);
  if (f.fix_code) console.log(`      ${f.fix_code}`);
}
for (const item of review.checklist) console.log(`  [${item.status}] ${item.item}: ${item.note}`);
for (const c of review.coverage_check) {
  console.log(`  ${c.id}: ${c.addressed ? "ok" : "SET ASIDE"} - ${c.note}`);
}

writeFileSync(review.rewrite.filename, review.rewrite.code);   // schema.refined.graphql
var started struct{ JobID string `json:"job_id"` }
if err := call("POST", "/run", payload, &started); err != nil {
	log.Fatal(err)
}

var job struct {
	Status string          `json:"status"`
	Error  string          `json:"error"`
	Output json.RawMessage `json:"output"`
}
for {
	if err := call("GET", "/jobs/"+started.JobID, nil, &job); err != nil {
		log.Fatal(err)
	}
	if job.Status == "succeeded" || job.Status == "failed" {
		break
	}
	time.Sleep(1500 * time.Millisecond)
}

// job.Output is {"output": "<json string>"} — unwrap, unquote, then unmarshal:
type Review struct {
	ReviewName   string `json:"review_name"`
	VerdictLevel string `json:"verdict_level"`
	Verdict      string `json:"verdict"`
	Health       []struct {
		Area, Status, Note string
	} `json:"health"`
	Findings []struct {
		Severity, Category, Title, Detail string
		FixCode                           string `json:"fix_code"`
	} `json:"findings"`
	Checklist []struct {
		Item, Status, Note string
	} `json:"checklist"`
	Rewrite struct {
		Filename, Code string
	} `json:"rewrite"`
}
var wrapper struct{ Output string `json:"output"` }
json.Unmarshal(job.Output, &wrapper)
var review Review
json.Unmarshal([]byte(wrapper.Output), &review)

fmt.Printf("%s [%s]: %s\n", review.ReviewName, review.VerdictLevel, review.Verdict)
for _, a := range review.Health {
	fmt.Printf("  [%s] %s: %s\n", a.Status, a.Area, a.Note)
}
for _, f := range review.Findings {
	fmt.Printf("  (%s) %s: %s\n", f.Severity, f.Category, f.Title)
}
for _, c := range review.Checklist {
	fmt.Printf("  [%s] %s: %s\n", c.Status, c.Item, c.Note)
}
os.WriteFile(review.Rewrite.Filename, []byte(review.Rewrite.Code), 0o644) // schema.refined.graphql
String envelope = api("POST", "/run", jsonPayload);
String jobId = /* data.job_id via your JSON library */;

while (true) {
    String job = api("GET", "/jobs/" + jobId, null);
    String status = /* data.status */;
    if (status.equals("succeeded") || status.equals("failed")) break;
    Thread.sleep(1500);
}
// The review is at data.output.output as a JSON string — parse it again, then read
// review_name, verdict_level, verdict, overview, health[] (five areas with area/status/note),
// findings[] (severity/category/title/detail/fix_code), checklist[] (item/status/note),
// coverage_check[] (id/addressed/note), rewrite{filename, code}, next_steps[] and summary.
// Finally write the refined code to disk:
//   Files.writeString(Path.of(rewriteFilename), rewriteCode);   // schema.refined.graphql
started = api("POST", "/run", payload)

job = nil
loop do
  job = api("GET", "/jobs/#{started["job_id"]}")
  break if %w[succeeded failed].include?(job["status"])
  sleep 1.5
end
raise (job["error"] || "run failed") if job["status"] == "failed"

raw = job["output"].is_a?(Hash) ? job["output"].fetch("output", job["output"]) : job["output"]
review = raw.is_a?(String) ? JSON.parse(raw) : raw

puts "#{review["review_name"]} [#{review["verdict_level"]}]: #{review["verdict"]}"
review["health"].each { |a| puts "  [#{a["status"]}] #{a["area"]}: #{a["note"]}" }
review["findings"].each do |f|
  puts "  (#{f["severity"]}) #{f["category"]}: #{f["title"]}"
  puts "      #{f["fix_code"]}" unless f["fix_code"].to_s.empty?
end
review["checklist"].each { |c| puts "  [#{c["status"]}] #{c["item"]}: #{c["note"]}" }
review["coverage_check"].each { |c| puts "  #{c["id"]}: #{c["addressed"] ? "ok" : "SET ASIDE"}" }

File.write(review["rewrite"]["filename"], review["rewrite"]["code"])   # schema.refined.graphql
$started = api("POST", "/run", $payload);

do {
    sleep(2);
    $job = api("GET", "/jobs/" . $started["job_id"]);
} while (!in_array($job["status"], ["succeeded", "failed"]));

if ($job["status"] === "failed") {
    throw new Exception($job["error"] ?? "run failed");
}

$raw = is_array($job["output"]) ? ($job["output"]["output"] ?? $job["output"]) : $job["output"];
$review = is_string($raw) ? json_decode($raw, true) : $raw;

echo "{$review['review_name']} [{$review['verdict_level']}]: {$review['verdict']}\n";
foreach ($review["health"] as $a) {
    echo "  [{$a['status']}] {$a['area']}: {$a['note']}\n";
}
foreach ($review["findings"] as $f) {
    echo "  ({$f['severity']}) {$f['category']}: {$f['title']}\n";
    if ($f["fix_code"] !== "") { echo "      {$f['fix_code']}\n"; }
}
foreach ($review["checklist"] as $item) {
    echo "  [{$item['status']}] {$item['item']}: {$item['note']}\n";
}
foreach ($review["coverage_check"] as $c) {
    echo "  {$c['id']}: " . ($c["addressed"] ? "ok" : "SET ASIDE") . "\n";
}

file_put_contents($review["rewrite"]["filename"], $review["rewrite"]["code"]);   // schema.refined.graphql
var started = await SkillSafe.ApiAsync(HttpMethod.Post, "/run", payload);
var jobId = started.GetProperty("job_id").GetString();

JsonElement job;
while (true)
{
    job = await SkillSafe.ApiAsync(HttpMethod.Get, $"/jobs/{jobId}");
    var status = job.GetProperty("status").GetString();
    if (status is "succeeded" or "failed") break;
    await Task.Delay(1500);
}

var rawText = job.GetProperty("output").GetProperty("output").GetString();
using var doc = JsonDocument.Parse(rawText!);
var review = doc.RootElement;

Console.WriteLine($"{review.GetProperty("review_name")} " +
                  $"[{review.GetProperty("verdict_level")}]: {review.GetProperty("verdict")}");
foreach (var a in review.GetProperty("health").EnumerateArray())
{
    Console.WriteLine($"  [{a.GetProperty("status")}] {a.GetProperty("area")}: {a.GetProperty("note")}");
}
foreach (var f in review.GetProperty("findings").EnumerateArray())
{
    Console.WriteLine($"  ({f.GetProperty("severity")}) {f.GetProperty("category")}: " +
                      $"{f.GetProperty("title")}");
}
foreach (var c in review.GetProperty("checklist").EnumerateArray())
{
    Console.WriteLine($"  [{c.GetProperty("status")}] {c.GetProperty("item")}: {c.GetProperty("note")}");
}

var rewrite = review.GetProperty("rewrite");
await File.WriteAllTextAsync(rewrite.GetProperty("filename").GetString()!,   // schema.refined.graphql
                             rewrite.GetProperty("code").GetString()!);

The model is asked for one JSON object and nothing else, but a stray code fence or preamble is always possible. Strip a leading ```json fence, take the text between the first { and the last }, and only then parse — that is what the app does before it falls back to a retry_note reformat run.

The review object — output schema

One JSON object, always the same shape. Every array is present (findings is empty only if genuinely nothing applies); health always has exactly the five areas, checklist always has exactly the twelve items, and rewrite.code is never empty. If the paste was too thin to review responsibly, you still get this object: what is there gets reviewed, the verdict says the paste is thin, and what you would need to show lands in next_steps. If the paste is not GraphQL at all, you still get the object — one high-severity finding explaining what arrived, every health area at risk, every checklist item at na, and a rewrite.code comment block saying what to paste instead. Resolver code in JavaScript, TypeScript, Python, Go or Ruby is reviewed too, at the level of its schema-contract and efficiency implications, and the review says it is doing so.

FieldTypeMeaning
review_namestringA short name for the review, taken from the schema's own type, operation or file naming.
verdict_levelstringsound (nothing material found), refactor (findings exist but are medium/low or only bite at scale) or rework (a high finding means the API is broken as pasted).
verdictstringOne or two sentences: the overall state and the single most important change.
overviewstringOne or two paragraphs: what this API exposes, and the pattern behind what was found.
healtharray of 5{area, status, note} — the five areas listed below, each exactly once. status is good (nothing material), risk (works, with caveats) or bad (a high-severity finding lives here). Each note references something concrete in the pasted code; an area with no evidence in the paste is risk, never good.
findingsarray{severity, category, title, detail, fix_code}. severity is high (a real schema-design violation or broken contract as pasted — an unbounded list field, an N+1 relationship resolver, ungated introspection on a public surface, authorization missing from a resolver) | medium (works today but degrades or scales badly) | low (polish); category is schema, nullability, pagination, resolvers, security, errors, naming or performance. detail quotes the type, field, argument or resolver it concerns; fix_code is corrected GraphQL SDL or resolver code in your own naming, or an empty string when the finding is a question or trade-off rather than a mechanical fix.
checklistarray of 12{item, status, note} — the twelve schema-design items listed below, each exactly once and in order. status is pass (the paste shows it handled), fail (the paste shows it mishandled — a finding backs this) or na (the paste gives no evidence either way). The note says what was seen or what is missing.
coverage_checkarray{id, addressed, note} — one entry per prescan_facts item you sent (ap:unpaginated-list, t:User, …), saying where the review covers it or why it was set aside (a keyword hit can be a false positive; the note says so). Nothing you flagged is silently dropped.
rewriteobject{filename, code}filename is normally schema.refined.graphql, and code is your own code refined: same domain, same intent, findings fixed — nullability tightened deliberately, list fields moved to connections, mutations given input and payload types, expected failures modelled as data, opaque ID scalars used, deprecations given reasons, enums introduced where bare strings were. Your naming, style and comments are preserved, and it is a complete replacement for what you pasted, not a fragment.
next_stepsstring[]Ordered and concrete: paginate User.posts, batch the author resolver with DataLoader, cap query depth, and so on.
summarystring3–5 sentences an API lead could paste into a review.

The five health areas, in order, spelled exactly like this:

areaWhat its note covers
Schema design & nullabilityTypes model the domain rather than the database, non-null is used where a value is genuinely guaranteed, optionality is a deliberate choice rather than a default, and enums and custom scalars stand in for bare strings.
Pagination & list shapeList fields are paginated — cursor connections with edges/pageInfo, or a documented offset scheme — rather than returning an unbounded array, and page sizes have a server-enforced maximum.
Resolver efficiencyRelationship resolvers batch through DataLoader instead of firing one query per parent, expensive fields are not resolved eagerly, and the resolver map does no work the selection set did not ask for.
Security & abuse limitsQuery depth and complexity are capped, introspection is gated on production surfaces, authorization is enforced inside resolvers rather than assumed at the gateway, and ids do not leak internal keys.
Errors & mutation contractsMutations take a single input and return a payload type, expected failures are modelled as data (error unions or a payload errors field) rather than thrown, and deprecations carry a reason.

The twelve checklist items, in order, spelled exactly like this:

itemWhat its note covers
Nullability designed intentionallyNon-null is applied where the value is genuinely always present, and a nullable field is a modelling decision rather than the default that fell out of the ORM.
List fields paginated with connectionsEvery list-returning field takes pagination arguments and returns a connection with pageInfo, instead of a bare [Type] that grows without bound.
Mutations use input and payload typesEach mutation takes one input argument and returns a named payload type, so arguments and results can evolve without breaking clients.
Expected failures modeled as dataBusiness failures come back in the payload — an error union or a typed errors field — rather than as thrown exceptions in the top-level errors array.
Relationship resolvers batched with DataLoaderFields that resolve per parent go through a batching loader, so a list of parents costs one query rather than one query per row.
Query depth and complexity limitedThe server rejects queries beyond a depth or complexity budget, so a recursive relationship cannot be turned into a denial of service.
Introspection gated in productionIntrospection and schema-dump endpoints are disabled or authenticated on public production surfaces, while staying available in development.
Authorization enforced in resolversEvery field that exposes protected data checks the caller in the resolver or via a directive, rather than trusting an upstream gateway or the shape of the query.
Consistent naming conventionsFields are camelCase, types PascalCase, enum values SCREAMING_SNAKE_CASE, and mutations read as verbs, so the schema is predictable to a client author.
IDs are opaque ID scalarsIdentifiers are typed ID and are opaque to clients, not raw database integers or strings that leak internal keys and row counts.
Deprecations carry a reasonEvery @deprecated field or enum value states why and what to use instead, so clients can migrate without asking.
Enums and scalars over bare stringsClosed sets are enums and formatted values are custom scalars (DateTime, URL, EmailAddress) rather than String with a comment.

A small, realistic result for the schema.graphql paste above, trimmed for length:

{
  "review_name": "schema.graphql - user and posts surface",
  "verdict_level": "rework",
  "verdict": "'User.posts' returns an unbounded list with no pagination and no batching, so
              a single query can walk the whole posts table row by row; paginate it and put
              a loader behind it before anything else.",
  "overview": "A two-type public surface: one root field 'user(id: ID!)' and a 'User' type
               that exposes 'posts'. The nullability is the default that fell out of the
               data layer rather than a decision - 'user' can return null with no way for a
               client to tell 'not found' from 'not allowed', and 'posts' is a nullable list
               of nullable elements. The list has no arguments at all, so there is no page
               size a server can enforce.",
  "health": [
    { "area": "Schema design & nullability", "status": "bad",
      "note": "'posts: [Post]' is a nullable list of nullable Posts with no stated reason;
               'user(id: ID!): User' returns null for both 'missing' and 'forbidden'." },
    { "area": "Pagination & list shape", "status": "bad",
      "note": "'posts' takes no 'first'/'after' arguments and returns a bare array, so the
               response size is bounded only by the size of the table." },
    { "area": "Resolver efficiency", "status": "risk",
      "note": "No resolver code was pasted, but 'User.posts' is the classic per-parent
               fetch; without a DataLoader a list of users is one query per user." },
    { "area": "Security & abuse limits", "status": "bad",
      "note": "Target is 'public' and the paste shows no depth or complexity cap, no
               introspection gate and no authorization on 'user'." },
    { "area": "Errors & mutation contracts", "status": "risk",
      "note": "No Mutation type is present, so the contract cannot be judged; the query
               side models failure only as a null." }
  ],
  "findings": [
    { "severity": "high", "category": "pagination",
      "title": "User.posts returns an unbounded list",
      "detail": "'posts: [Post]' has no arguments, so a client can ask for every post a
                 user has ever written in one field selection.",
      "fix_code": "type User {\n  id: ID!\n  posts(first: Int! = 20, after: String): PostConnection!\n}\n\ntype PostConnection {\n  edges: [PostEdge!]!\n  pageInfo: PageInfo!\n}" },
    { "severity": "high", "category": "resolvers",
      "title": "The posts relationship will fan out one query per user",
      "detail": "'User.posts' resolves per parent; any query returning a list of users
                 multiplies into N post queries.",
      "fix_code": "const postsByUser = new DataLoader(ids =>\n  db.posts.byUserIds(ids)\n);\n\nUser: {\n  posts: (user, args, ctx) => ctx.loaders.postsByUser.load(user.id),\n}" },
    { "severity": "high", "category": "security",
      "title": "A public surface with no depth or complexity budget",
      "detail": "'target' is 'public' and nothing caps recursion; 'user.posts.author.posts'
                 repeats as deep as the client cares to type.",
      "fix_code": "validationRules: [\n  depthLimit(8),\n  complexityLimit({ maximumComplexity: 1000 }),\n]" },
    { "severity": "medium", "category": "nullability",
      "title": "posts is a nullable list of nullable elements",
      "detail": "'[Post]' lets both the list and each element be null, so every client must
                 write two null checks for a case the server never intends to produce.",
      "fix_code": "posts(first: Int! = 20, after: String): PostConnection!" },
    { "severity": "medium", "category": "errors",
      "title": "A null user cannot distinguish missing from forbidden",
      "detail": "'user(id: ID!): User' returns null for a deleted user, an unknown id and an
                 unauthorized caller alike.",
      "fix_code": "union UserResult = User | NotFound | Forbidden\n\ntype Query {\n  user(id: ID!): UserResult!\n}" },
    { "severity": "low", "category": "schema",
      "title": "No custom scalars for formatted values",
      "detail": "The Post type referenced here is not shown, but the surface declares no
                 DateTime or URL scalar, so timestamps will arrive as String.",
      "fix_code": "scalar DateTime\n\ntype Post {\n  publishedAt: DateTime\n}" }
  ],
  "checklist": [
    { "item": "Nullability designed intentionally", "status": "fail",
      "note": "'[Post]' is nullable at both levels with no stated reason." },
    { "item": "List fields paginated with connections", "status": "fail",
      "note": "'posts' takes no pagination arguments and returns a bare list." },
    { "item": "Mutations use input and payload types", "status": "na",
      "note": "No Mutation type in the paste." },
    { "item": "Expected failures modeled as data", "status": "fail",
      "note": "A missing or forbidden user is signalled only by a null." },
    { "item": "Relationship resolvers batched with DataLoader", "status": "na",
      "note": "No resolver code was pasted; the SDL alone cannot show batching." },
    { "item": "Query depth and complexity limited", "status": "na",
      "note": "Server configuration was not included in the paste." },
    { "item": "Introspection gated in production", "status": "na",
      "note": "Server configuration was not included in the paste." },
    { "item": "Authorization enforced in resolvers", "status": "na",
      "note": "No resolver code was pasted." },
    { "item": "Consistent naming conventions", "status": "pass",
      "note": "camelCase fields and PascalCase types throughout." },
    { "item": "IDs are opaque ID scalars", "status": "pass",
      "note": "'User.id' and the 'user(id:)' argument are both the ID scalar." },
    { "item": "Deprecations carry a reason", "status": "na",
      "note": "Nothing is deprecated in the paste." },
    { "item": "Enums and scalars over bare strings", "status": "na",
      "note": "No enum-shaped or formatted fields appear on the two types shown." }
  ],
  "coverage_check": [
    { "id": "ap:unpaginated-list", "addressed": true,
      "note": "Covered by the pagination finding - 'posts' moves to a PostConnection." },
    { "id": "t:User", "addressed": true,
      "note": "The type under review; refined in full in rewrite.code." }
  ],
  "rewrite": { "filename": "schema.refined.graphql",
               "code": "type Query {\n  user(id: ID!): UserResult!\n}\n\ntype User { … }" },
  "next_steps": [
    "Give 'User.posts' 'first'/'after' arguments and a PostConnection return type.",
    "Put a DataLoader behind the posts resolver so a list of users costs one query.",
    "Add a depth limit and a complexity budget, and gate introspection on production.",
    "Return a UserResult union so clients can tell missing from forbidden."
  ],
  "summary": "The types are named well and use ID scalars, but the list field is unbounded. …"
}

The refined rewrite is a starting point, not a sign-off: it is written to be complete and self-consistent with the findings, but it is AI-generated and it only sees what you pasted. Read it, run it through your schema linter and composition check, replay your clients' real queries against it, and keep the human review in the loop before it goes anywhere near production — a schema change is a contract change.

Step 5 — Stream the review as it is written

POST /run-stream

/run-stream takes exactly the same body as /run but answers with server-sent events, so you can show progress instead of a spinner — useful here because the refined rewrite makes for a long reply. This app's own progress panel is this endpoint. Events are separated by a blank line; each has an event: line and a data: line carrying JSON.

EventPayloadMeaning
job{job_id, status}Sent once, when the job is accepted — show "starting".
delta{text}A chunk of the reply, in order. Append it; the accumulated length is your only progress signal (the total is not known in advance).
done{job_id, status, charged_credits, output}The final, authoritative result — read the review from output.output rather than trusting concatenated deltas, and the settled price from charged_credits.
error{code, message}Replaces done when the run fails.
# -N disables buffering so events print as they arrive
curl -N -s -X POST "$API/run-stream" \
  -H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \
  -H "Idempotency-Key: review-$(date +%s)" \
  -d @input.json

# event: job
# data: {"job_id":"job_...","status":"running"}
#
# event: delta
# data: {"text":"{\"review_name\":\"schema.graphql"}
# ...
# event: done
# data: {"job_id":"job_...","status":"succeeded","charged_credits":612,"output":{"output":"{...}"}}
import json, requests

result = None
with requests.post(
    API + "/run-stream",
    headers={"Authorization": f"Bearer {TOKEN}",
             "Idempotency-Key": "review-001"},
    json=payload,
    stream=True,
) as r:
    r.raise_for_status()
    event = None
    for line in r.iter_lines(decode_unicode=True):
        if not line:
            continue
        if line.startswith("event:"):
            event = line[len("event:"):].strip()
        elif line.startswith("data:"):
            data = json.loads(line[len("data:"):].strip())
            if event == "delta":
                print(".", end="", flush=True)          # live progress
            elif event == "done":
                result = data
            elif event == "error":
                raise RuntimeError(data.get("message", "run failed"))

review = json.loads(result["output"]["output"])         # authoritative
print("charged:", result["charged_credits"], "-", review["review_name"])
for area in review["health"]:
    print(f'  [{area["status"]}] {area["area"]}')
open(review["rewrite"]["filename"], "w", encoding="utf-8").write(review["rewrite"]["code"])
const res = await fetch(API + "/run-stream", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${TOKEN}`,
    "Content-Type": "application/json",
    "Idempotency-Key": crypto.randomUUID(),
  },
  body: JSON.stringify(payload),
});

const reader = res.body.getReader();
const decoder = new TextDecoder();
let buf = "", done = null;

for (;;) {
  const chunk = await reader.read();
  if (chunk.done) break;
  buf += decoder.decode(chunk.value, { stream: true });
  const frames = buf.split("\n\n");
  buf = frames.pop();
  for (const frame of frames) {
    const name = /^event:\s*(.+)$/m.exec(frame)?.[1];
    const body = /^data:\s*(.+)$/m.exec(frame)?.[1];
    if (!name || !body) continue;
    const data = JSON.parse(body);
    if (name === "delta") process.stdout.write(".");   // live progress
    if (name === "done") done = data;
    if (name === "error") throw new Error(data.message ?? "run failed");
  }
}

const review = JSON.parse(done.output.output);
console.log(`\n${done.charged_credits} credits - ${review.review_name}`);
for (const area of review.health) console.log(`  [${area.status}] ${area.area}`);
writeFileSync(review.rewrite.filename, review.rewrite.code);   // schema.refined.graphql
body, _ := json.Marshal(payload)
req, _ := http.NewRequest("POST", API+"/run-stream", bytes.NewReader(body))
req.Header.Set("Authorization", "Bearer "+token)
req.Header.Set("Content-Type", "application/json")
req.Header.Set("Idempotency-Key", "review-001")

res, err := http.DefaultClient.Do(req)
if err != nil {
	log.Fatal(err)
}
defer res.Body.Close()

var event string
var final map[string]any
sc := bufio.NewScanner(res.Body)
sc.Buffer(make([]byte, 0, 64*1024), 4*1024*1024)
for sc.Scan() {
	line := sc.Text()
	switch {
	case strings.HasPrefix(line, "event:"):
		event = strings.TrimSpace(strings.TrimPrefix(line, "event:"))
	case strings.HasPrefix(line, "data:"):
		var data map[string]any
		json.Unmarshal([]byte(strings.TrimPrefix(line, "data:")), &data)
		switch event {
		case "delta":
			fmt.Print(".") // live progress
		case "done":
			final = data
		case "error":
			log.Fatal(data["message"])
		}
	}
}
// final["output"].(map[string]any)["output"].(string) is the review JSON —
// unmarshal it into the Review struct from step 4, then write review.Rewrite.Code to disk.
// Java 17+ — read the stream line by line instead of buffering the body.
var req = HttpRequest.newBuilder(URI.create(API + "/run-stream"))
    .header("Authorization", "Bearer " + TOKEN)
    .header("Content-Type", "application/json")
    .header("Idempotency-Key", "review-001")
    .POST(HttpRequest.BodyPublishers.ofString(jsonPayload))
    .build();

var res = HTTP.send(req, HttpResponse.BodyHandlers.ofLines());
String event = null, done = null;
for (String line : (Iterable<String>) res.body()::iterator) {
    if (line.startsWith("event:")) {
        event = line.substring(6).trim();
    } else if (line.startsWith("data:")) {
        String data = line.substring(5).trim();
        if ("delta".equals(event)) System.out.print(".");   // live progress
        else if ("done".equals(event)) done = data;
        else if ("error".equals(event)) throw new RuntimeException(data);
    }
}
// parse `done`, then parse data.output.output again — it is a JSON string holding
// review_name, verdict_level, health[], findings[], checklist[], rewrite{filename, code} and the rest.
require "net/http"
require "json"

uri = URI(API + "/run-stream")
req = Net::HTTP::Post.new(uri)
req["Authorization"] = "Bearer #{TOKEN}"
req["Content-Type"] = "application/json"
req["Idempotency-Key"] = "review-001"
req.body = payload.to_json

event = nil
done = nil
Net::HTTP.start(uri.host, uri.port, use_ssl: true) do |http|
  http.request(req) do |res|
    res.read_body do |chunk|
      chunk.each_line do |line|
        line = line.strip
        if line.start_with?("event:")
          event = line.delete_prefix("event:").strip
        elsif line.start_with?("data:")
          data = JSON.parse(line.delete_prefix("data:").strip)
          case event
          when "delta" then print "."           # live progress
          when "done"  then done = data
          when "error" then raise (data["message"] || "run failed")
          end
        end
      end
    end
  end
end

review = JSON.parse(done["output"]["output"])
puts "\n#{done["charged_credits"]} credits - #{review["review_name"]}"
review["health"].each { |a| puts "  [#{a["status"]}] #{a["area"]}" }
File.write(review["rewrite"]["filename"], review["rewrite"]["code"])   # schema.refined.graphql
$event = null;
$done  = null;

$ch = curl_init(API . "/run-stream");
curl_setopt_array($ch, [
    CURLOPT_POST       => true,
    CURLOPT_HTTPHEADER => [
        "Authorization: Bearer $TOKEN",
        "Content-Type: application/json",
        "Idempotency-Key: review-001",
    ],
    CURLOPT_POSTFIELDS => json_encode($payload),
    CURLOPT_WRITEFUNCTION => function ($ch, $chunk) use (&$event, &$done) {
        foreach (explode("\n", $chunk) as $line) {
            $line = trim($line);
            if (str_starts_with($line, "event:")) {
                $event = trim(substr($line, 6));
            } elseif (str_starts_with($line, "data:")) {
                $data = json_decode(trim(substr($line, 5)), true);
                if ($event === "delta") { echo "."; }        // live progress
                elseif ($event === "done") { $done = $data; }
                elseif ($event === "error") { throw new Exception($data["message"] ?? "run failed"); }
            }
        }
        return strlen($chunk);
    },
]);
curl_exec($ch);
curl_close($ch);

$review = json_decode($done["output"]["output"], true);
echo "\n{$done['charged_credits']} credits - {$review['review_name']}\n";
foreach ($review["health"] as $a) { echo "  [{$a['status']}] {$a['area']}\n"; }
file_put_contents($review["rewrite"]["filename"], $review["rewrite"]["code"]);   // schema.refined.graphql
var req = new HttpRequestMessage(HttpMethod.Post, Api + "/run-stream") {
    Content = JsonContent.Create(payload),
};
req.Headers.Add("Idempotency-Key", "review-001");

using var res = await Http.SendAsync(req, HttpCompletionOption.ResponseHeadersRead);
using var reader = new StreamReader(await res.Content.ReadAsStreamAsync());

string? evt = null, done = null;
while (await reader.ReadLineAsync() is { } line)
{
    if (line.StartsWith("event:")) evt = line[6..].Trim();
    else if (line.StartsWith("data:"))
    {
        var data = line[5..].Trim();
        if (evt == "delta") Console.Write(".");            // live progress
        else if (evt == "done") done = data;
        else if (evt == "error") throw new Exception(data);
    }
}

using var final = JsonDocument.Parse(done!);
var text = final.RootElement.GetProperty("output").GetProperty("output").GetString();
using var reviewDoc = JsonDocument.Parse(text!);
var review = reviewDoc.RootElement;
Console.WriteLine(review.GetProperty("review_name"));
foreach (var a in review.GetProperty("health").EnumerateArray())
    Console.WriteLine($"  [{a.GetProperty("status")}] {a.GetProperty("area")}");
var rewrite = review.GetProperty("rewrite");
await File.WriteAllTextAsync(rewrite.GetProperty("filename").GetString()!,   // schema.refined.graphql
                             rewrite.GetProperty("code").GetString()!);

In a browser, the native EventSource only speaks GET, and this endpoint is a POST — read the fetch response body incrementally, as the JavaScript sample above does. On an idempotent replay the server may answer with a plain JSON envelope instead of an event stream; check the Content-Type before you start parsing frames.