B. Quality, Speed & Resilience · Prompt 13

Why it matters

An app that feels instant in one browser tab can stall when many people use it at once.

Modeled on

Grafana k6, a load-testing tool.

How to run this prompt

1. Switch to a mode that does not edit files. In Cursor that is Ask or Plan. In Claude Code that is Plan mode.
2. Paste the audit prompt. Wait for the report. It must stop and ask which IDs to fix.
3. Read the report. Keep the IDs you agree with.
4. Switch to a mode that can edit. Paste the fix prompt and the IDs you chose.
5. Switch back to the read-only mode and paste the same audit prompt again. Confirm those IDs are gone.

• Cursor: audit in Ask mode or Plan mode. Fix in Agent mode.
• Claude Code: audit in Plan mode (Shift+Tab cycles to it). Fix in Normal mode, which can edit.
• Any other tool: audit in Chat, Discuss, or Plan mode, whichever answers without editing files. If the tool has no such mode, the prompt itself forbids edits. Fix in the mode that is allowed to edit files.

MODE: AUDIT ONLY. Do not create, edit, or delete any file. Do not run commands
that change anything: no installs, migrations, git commits, deploys, or "--fix" flags.
If your tool has an Ask, Plan, Chat, or Discuss mode, use it for this prompt.

Before you start:
- Tell me the stack you detect (framework, language, database, auth, hosting,
  payment provider) and which folders you will review.
- If a check below does not apply to this stack, write "Not applicable" and why.
- If you can run read-only commands, run the ones listed. If you cannot, list them
  so I can run them and paste the output.

Performance & Load Simulation: what to check
Review this code as if it will be hit by many concurrent users. Do not run a load test against production. You may describe a k6 script. Do not start one unless I give you a staging URL.
1. Find handlers that read or write a table with no pagination and no limit. Cite the query.
2. Find blocking work inside a request: a slow external call, a large file read, or heavy CPU, with no timeout and no queue.
3. Find database access that opens a new connection per request instead of using a pool. Say what the driver config shows.
4. Find handlers called often with the same inputs and no cache. Do not invent a cache product. Say what could be cached and for how long.
5. Find loops that query or call an API once per item (N+1) where one batched query would do.
6. Find request-specific state stored in a module-level variable or memory. That breaks when a second instance serves the next request.
7. For each issue, name the file, why it gets worse as concurrency grows, and a rough guess of when it would hurt. Label the guess as a guess.
8. List the read-only queries or EXPLAIN plans that would confirm the worst items. Do not run EXPLAIN ANALYZE on a statement that writes.

Evidence rules:
- Every finding cites a file path and line number, or the exact command output used.
- Mark each finding Confirmed (seen in the code) or Needs manual check (depends on
  something outside the repo, such as a dashboard setting or production data).
- Never print a full secret. Show the first 4 characters and the location only.
- If you are not sure, say so. Do not invent files, settings, or results.

Severity: Critical = exploitable now, or leaks real data or money. High = serious
with little effort. Medium = weakens defenses or needs a second bug. Low = hygiene.

Report:
- Summary: count of findings by severity.
- Table: ID | Severity | Confirmed? | Finding | Evidence | Why it matters | Suggested fix | Effort
  (IDs for this prompt use the prefix P13, for example P13-1, P13-2.)
- Checked and fine: what you verified is already OK.
- Could not check: what I need to look at myself, and where.

Then stop. Do not fix anything. Ask me which IDs I want fixed.

Related prompts

• 14. Slow Network, Offline & API Failure Handling

Share this prompt