Prompt injection
Related service Cybersecurity
DEFINITION
Attack where user input overrides the LLM system prompt. Five categories: direct, indirect, RAG-poisoning, tool-abuse, exfiltration. OWASP LLM Top-1.
- Threat model→
A structured exercise that walks the system's actors, attack surface, risks, and controls. Day one of every DField project · before any code.
- Penetration test (pentest)→
Manual + tooled attack simulation that reveals what an attacker could achieve. We deliver findings as PRs in your repo, not an 80-page PDF.
- DevSecOps→
Security as a continuously-running CI step (SAST, DAST, SCA, IaC scan), not an annual project. Runs against every push; every sprint closes at least one security bug.
- MFA (Multi-factor auth)→
Two or more factors (TOTP, WebAuthn, biometric) beyond a password. Table-stakes in SaaS today · enterprise procurement disqualifies you without it.
- SOC 2→
A US audit framework for confidentiality, integrity, availability, and privacy controls. For SaaS, the Type II audit (6-12 months of observation) is the standard enterprise baseline.
- ISO 27001→
International standard for Information Security Management Systems (ISMS). Often preferred in Europe instead of or alongside SOC 2. 3-year certification cycle.
- 0130 Sep 2026Q3 2026 roundup: what shifted, what we shipped, what broke→
- 0201 Jul 2026Q2 2026 roundup: what shifted, what we shipped, what broke→
- 0302 Jun 2026AI agent pricing 2026: what an autonomous agent costs→
- 0402 Jun 2026H1 2026 in review: what changed for EU software teams→
- 0502 Jun 2026AI in logistics & supply chain: 2026 SME guide→
- 0614 May 2026The EU AI Act in practice: a 2026 guide for AI teams→
- 0714 May 2026The 2026 smart contract security checklist before you ship→
- 0814 May 2026Self-hosted AI or the API? When to run your own LLM in 2026→