Five Silent Assumptions That Turn AI Code Into Debt

A developer has cataloged five silent assumptions that turn AI-generated code into technical debt, treating them as merge blockers. The anti-patterns include invented environment variables, unvetted imports, missing auth, incorrect schema references, and hidden side effects. The developer proposes a constraints file and gate script to enforce rules on staged diffs.

Generated code looks cheap until the assumptions land. You own every invented config, schema, and side effect. I now treat silent model assumptions as merge blockers. Did the model ask about your auth scheme? Did it confirm the table names you actually have? If not, you did not get a patch. You got a guess. This is a catalog, not a pep talk. Each anti-pattern has a symptom, a root cause, and a replacement. Then I give you a constraint file and a gate script you can run on a staged diff. Agents fill gaps because that is their job. Gap filling helps during a short spike. It becomes poison inside a brownfield service. Cheap tokens make the guesses faster. They do not make the guesses true. Technical debt arrives as confident Python. I keep a constraints file next to the repo. The model must read it before it writes. A small gate fails the patch when it invents facts. Symptom The patch reads PAYMENT WEBHOOK SECRET on boot. Your secrets manager has never heard of it. Staging dies inside os.environ ... . Root cause The model completed a typical tutorial shape. Tutorials always hide one missing env var. Nobody listed the real allowlist. Replacement Publish an env allowlist in the repo. Reject new keys unless a human adds them. constraints.yml env allowlist: - DATABASE URL - REDIS URL - APP ENV forbidden env prefixes: - AWS - STRIPE - OPENAI Ask yourself: who named that variable? If the answer is "the model," delete it. Symptom requirements.txt gains httpx or orjson . Sometimes the name does not even exist. CI installs it, or CI cannot install it. Root cause The model optimized for a blog-post stack. It did not read your lockfile. Convenience beat your supply-chain rules. Replacement Diff imports against the lockfile. Unknown import roots fail the gate. Humans add libraries on purpose. ALLOWED IMPORT ROOTS = {"flask", "sqlalchemy", "redis", "pydantic"} Would you merge a mystery wheel from the internet? Then do not merge a mystery import either. Symptom The new endpoint has no auth decorator. Or it checks a header the gateway never sets. Or it trusts user id from the JSON body. Root cause Demos skip auth to keep the snippet small. The model learned those demos. Your threat model never traveled in the prompt. Replacement State the auth contract in constraints. Every HTTP handler must match one pattern. No pattern, no merge. http: must use decorator: "require session" forbid body fields as identity: - user id - account id - is admin Can an anonymous caller hit this route? If you cannot answer, the patch is incomplete. Symptom The query selects users.uuid . Your table has users.id . Or the patch adds metadata JSON nobody migrated. Root cause Language models remember popular schemas. They do not remember yours. A plausible column is still a lie. Replacement Check identifiers against a schema dump. I keep schema/tables.txt generated from migrations. Unknown columns fail the same way unknown env fails. schema/tables.txt generated, not hand-waved users.id users.email users.created at orders.id orders.user id orders.total cents Did you run the migration, or did the model imagine it? Imagination is not a migration. Symptom The helper writes /tmp/cache.json . It shells out to curl . It logs access tokens at INFO. Root cause The model "finished" the function. Finishing is not the same as isolating. Side effects feel like completeness. Replacement Ban whole families of calls in generated diffs. Allow them only in named modules. Keep the blast radius tiny. side effects: forbid substrings: - "subprocess." - "os.system " - "pathlib.Path '/tmp" - "open '/tmp" forbid log names: - password - token - authorization If a spike needs /tmp , put it in scratch/ . Do not let it ride into app/ . Here is a compact patch I would reject on sight. It looks helpful. It is five anti-patterns in one function. python + @app.route "/refunds", methods= "POST" + def refunds : + key = os.environ "STRIPE KEY" + import requests + user id = request.json "user id" + row = db.execute "SELECT uuid FROM users WHERE id=%s", user id + open "/tmp/refunds.log", "a" .write str request.json + return {"ok": True, "user id": user id} What did the model invent? A secret name. A new HTTP client. Identity from the body. A column you do not have. A world-readable temp log. The gate below should print failures, not a green check. If it passes this diff, your allowlists are too wide. Here is a small checker you can copy. It is a heuristic, not a full program analysis. Label it as a merge gate, not a proof. Save constraints.yml at the repo root. Save this script as tools/assumption gate.py . Feed it a unified diff from the model. bash /usr/bin/env python3 """Fail a generated diff that invents facts. This is a proposed gate. Run it on your own diffs. It does not execute the patch. It only scans added text. """ from future import annotations import argparse import re import sys from pathlib import Path import yaml ENV RE = re.compile r"os\.environ ?:\ |\.get\ '\" A-Z0-9 + " IMPORT RE = re.compile r"^ ?:from|import \s+ a-zA-Z0-9 \. + ", re.M IDENT RE = re.compile r"\b a-z a-z0-9 \. a-z a-z0-9 \b" ROUTE RE = re.compile r"@app\.route|@router\." def load constraints path: Path - dict: data = yaml.safe load path.read text if not isinstance data, dict : raise ValueError "constraints.yml must be a mapping" return data def added lines diff text: str - str: lines = for line in diff text.splitlines : if line.startswith "+" and not line.startswith "+++" : lines.append line 1: return "\n".join lines def main - int: parser = argparse.ArgumentParser parser.add argument "--diff", required=True parser.add argument "--constraints", default="constraints.yml" parser.add argument "--schema", default="schema/tables.txt" args = parser.parse args constraints = load constraints Path args.constraints diff = Path args.diff .read text encoding="utf-8" added = added lines diff failures: list str = allow env = set constraints.get "env allowlist", for key in ENV RE.findall added : if key not in allow env: failures.append f"phantom env: {key}" allowed imports = set constraints.get "allowed import roots", for raw in IMPORT RE.findall added : root = raw.split "." 0 if allowed imports and root not in allowed imports: failures.append f"unapproved import: {root}" schema path = Path args.schema if schema path.exists : allowed cols = { tuple line.strip .split ".", 1 for line in schema path.read text .splitlines if "." in line } known tables = {table for table, in allowed cols} for table, col in IDENT RE.findall added : if table in known tables and table, col not in allowed cols: failures.append f"invented column: {table}.{col}" for blob in constraints.get "side effects", {} .get "forbid substrings", : if blob in added: failures.append f"side effect: {blob r}" decorator = constraints.get "http", {} .get "must use decorator" if decorator and ROUTE RE.search added and decorator not in added: failures.append "route without auth decorator" identity fields = constraints.get "http", {} .get "forbid body fields as identity", for field in identity fields: if re.search rf"json\ '" {field} '" \ ", added : failures.append f"body used as identity: {field}" if not failures: print "assumption gate: pass" return 0 print "assumption gate: fail" for item in failures: print f" - {item}" return 1 if name == " main ": sys.exit main Run it like this: git diff --staged /tmp/staged.diff python tools/assumption gate.py --diff /tmp/staged.diff No staged diff? Pipe the model output through diff -u /dev/null . The gate still sees every added line. That is enough to catch the five patterns above. Expected output on the refunds example: assumption gate: fail - phantom env: STRIPE KEY - unapproved import: requests - invented column: users.uuid - side effect: "open '/tmp" - route without auth decorator - body used as identity: user id If that list is empty, the gate is not wired. Fix the constraints before you blame the model. | If the diff... | Treat it as | Human action | |---|---|---| | Adds an env key | Phantom config | Add to allowlist or delete | | Adds an import root | Unapproved dependency | Lockfile first, then code | | Adds a route, no decorator | Happy-path auth | Wrap or reject | Uses unknown table.col | Invented schema | Dump schema, then rewrite | Touches /tmp or subprocess | Invisible side effect | Move to scratch/ or drop | Print this table in the PR template. Reviewers stop arguing taste. They argue facts. Do not ask for "a refunds endpoint." Ask for a diff that obeys the file. Keep the prompt boring and strict. Read constraints.yml and schema/tables.txt. Return a unified diff only. Do not add env keys outside env allowlist. Do not add import roots outside allowed import roots. Do not invent columns. Every new route must use require session. If a fact is missing, ask a question. Do not guess. Then paste the gate failures back. The second turn should shrink, not sprawl. If it sprawls, the model is still filling gaps. Stop and edit by hand. I want the model to propose code. I do not want it to propose reality. Those are different jobs. Disclosure: This article was prepared as part of MonkeyCode's product outreach. I run the generate-then-gate loop on MonkeyCode's free models and free server. The server is a lab for the checker, not a factory for production traffic. The loop is boring on purpose: assumption gate.py .Free model access makes the retries cheap. The gate makes the retries honest. Without the gate, cheap retries just multiply debt. This gate reads text. It does not run tests. It will miss a renamed import alias. It will miss SQL built with f-strings. Object attributes look like columns. user.email can false-positive when users.email is real. Keep the schema file on table names, not on instance names. Review those hits instead of auto-fixing them. It can also nag on legitimate new columns. That is the point of a human allowlist. A noisy fail is better than a silent schema lie. Do not call this a security audit. Do not skip unit tests because the gate passed. Do not point the lab server at production databases. Regex will rot as your framework changes. Budget an hour when you upgrade the web layer. Update the decorator name. Update the import roots. Skip this if you have no lockfile. Skip this if you cannot dump schema. Skip this if the repo is a throwaway spike. Also skip it for generated front-end CSS churn. The patterns above target service code. A linter war on class names helps nobody. If you cannot review the allowlists, stop. An outdated allowlist becomes a rubber stamp. Rubber stamps are how assumptions sneak back. I want candidate diffs. I want them small. I want every new fact to be named. Ask the model: which constraints did you use? Ask it: which facts did you invent anyway? If it cannot list them, distrust the patch. Cheap code is a throughput trick. Assumption control is the actual engineering. Keep the catalog next to the gate, not in a wiki.