Command Reference
1. Test a Saved Burp Request File Safely
Load raw HTTP headers, cookies, and POST body from a file and test a single parameter.
sqlmap -r request.txt -p id --batch --level=2 --risk=1 --banner
2. Verify Current Database & User Only
Minimal proof-of-concept query without dumping sensitive customer tables.
sqlmap -r request.txt -p id --batch --current-db --current-user
3. Specify Backend DBMS Explicitly
Skip broad fingerprinting when you already know the database engine (postgresql, mysql, mssql).
sqlmap -r request.txt -p id --dbms=postgresql --technique=BEU --batch
4. List Tables in a Specific Schema
Enumerate table names for an authorized database schema.
sqlmap -r request.txt -D app_production --tables --batch
5. Throttle Requests & Route Through Burp Proxy
Add a 1-second delay between requests and inspect all traffic in Burp Suite.
sqlmap -r request.txt -p id --proxy="http://127.0.0.1:8080" --delay=1 --threads=1
6. Test JSON or GraphQL POST Payloads
Mark the exact injection point inside complex JSON with an asterisk (*).
sqlmap -u "https://example.com/api/search" --data='{"query":"test*"}' -H "Content-Type: application/json" --batch
7. Use Random User-Agent & Flush Session Cache
Start a clean test run with a standard browser User-Agent header.
sqlmap -r request.txt -p id --random-agent --flush-session --batch
8. Time-Based Blind SQLi Tuning
Increase --time-sec on high-latency links to avoid false positives.
sqlmap -r request.txt -p id --technique=T --time-sec=5 --batch
Frequently Asked Questions
What do the letters in --technique=BEUSTQ mean?
B = Boolean-based blind, E = Error-based, U = UNION query-based, S = Stacked queries, T = Time-based blind, and Q = Inline queries.
Why should pentesters keep --risk=1 on production-like systems?
Higher risk levels (--risk=3) include OR-based payloads that can evaluate to TRUE across an entire table, potentially altering every row if the vulnerable query happens to be an UPDATE or DELETE statement.