- Home
- LLM Security Tools
- Prompt Inspector
Prompt Inspector
What it is
Prompt Inspector provides prompt injection detection, sensitive word filtering, and AI safety guardrails through an API call. Its engine checks for prompt injections, jailbreaks, and system prompt leaks using layers that include hash caching, sensitive word matching, semantic vector analysis, optional AI review, and arbitration. Developers can use a REST API, Python or Node.js SDK, MCP server, and agent skills for OpenClaw and Claude Code. Cursor, VS Code Copilot, and Claude Desktop are identified as MCP clients. Each app can have its own API key, sensitive word settings, rate limits, and detection logs. The free plan includes 1,000 detections per month. Cloud pricing is described as pay-as-you-grow, but no price is stated. An API article lists Free limits of 10,000 monthly requests and 10 requests per second, distinct from the homepage detection allowance. Self-hosting is offered for non-commercial use under AGPL-3.0; enterprise includes private or on-premise deployment, custom integrations, managed payload library updates, and dedicated SLA support.
For and against
For
- Checks for injections, jailbreaks, and prompt leaks.
- Offers REST API and Python and Node.js SDKs.
- Per-app keys, settings, rate limits, and logs.
- Free plan includes 1,000 monthly detections.
- Self-hosted option available for non-commercial use.
Against
- Cloud pricing has no stated price.
- Free allowance differs between homepage and API article.
- Self-hosting is non-commercial under AGPL-3.0.
The verdict
Prompt Inspector combines several detection layers with developer integrations and per-app controls. Check the differing Free limits and unstated cloud pricing against your expected use.
Who it suits
It suits developers and teams adding prompt safety checks to applications through APIs, SDKs, or MCP. The self-hosted option is for non-commercial use, while enterprise customers are offered private or on-premise deployment and dedicated SLA support.
Plans and prices
Free
1,000 detections/month · no credit card stated
Read 15 Oct 2026 · appsecsanta.com
Enterprise
Private/on-premises deployment · dedicated support
Read 15 Oct 2026 · appsecsanta.com
Pro
Cloud service; exact allowance and price not stated
Read 15 Oct 2026 · appsecsanta.com
Starter
100,000 requests/month · 50 req/s · < 50 ms P95
Read 4 Oct 2026 · promptinspector.io
On the record
- Agent skills
- The integration repository includes zero-dependency Python and Node scripts for OpenClaw, Claude Code and other skill-capable agents.github.com · read 15 Oct 2026
- AI review
- Ambiguous inputs can be escalated to DeepSeek, Qwen or Google Gemini for an additional verdict.github.com · read 15 Oct 2026
- API limits
- The maker's API article lists 10,000 monthly requests and 10 requests per second for Free, while the homepage lists 1,000 monthly detections.promptinspector.io · read 4 Oct 2026
- Authentication
- SDKs accept an API key directly or through the PMTINSP_API_KEY environment variable, while MCP accepts X-App-Key or Bearer credentials.github.com · read 15 Oct 2026
- Custom filtering
- Each tenant can define its own sensitive words and regular-expression rules in addition to the built-in threat library.docs.promptinspector.io · read 15 Oct 2026
- Detection funnel
- Its pipeline combines SHA-256 caching, Aho-Corasick keyword and regex checks, vector similarity, optional LLM review and score arbitration.github.com · read 15 Oct 2026
- Detection layers
- The engine uses hash caching, sensitive word matching, semantic vector analysis, optional AI review, and arbitration.promptinspector.io · read 4 Oct 2026
- Documentation depth
- The documentation site covers architecture, REST use, Python and Node SDKs, MCP setup and open-source deployment.docs.promptinspector.io · read 15 Oct 2026
- Enterprise
- The enterprise offering includes private or on-premise deployment, custom integrations, managed payload library updates, and dedicated SLA support.promptinspector.io · read 4 Oct 2026
- Enterprise edition
- Enterprise adds analytics and audit logs, team roles, SSO/SAML, SLA-backed support, onboarding and custom model fine-tuning.github.com · read 15 Oct 2026
- Free allowance
- The homepage states the free plan includes 1,000 detections per month.promptinspector.io · read 4 Oct 2026
- Independent architecture praise
- AppSecSanta praises the five-stage funnel because inexpensive checks handle obvious threats before slower analysis is used.appsecsanta.com · read 15 Oct 2026
- Independent cost claim
- AppSecSanta says the staged design is intended to cost 60–95% less and run 2–7 times faster than LLM-only detection.appsecsanta.com · read 15 Oct 2026
- Independent integration praise
- AppSecSanta highlights REST, Python, Node.js and MCP support as fitting common development and production workflows.appsecsanta.com · read 15 Oct 2026
- Independent performance claim
- AppSecSanta reports typical response times of roughly 10 to 42 milliseconds and near-zero latency for cached threats.appsecsanta.com · read 15 Oct 2026
- Independent self-hosting fact
- AppSecSanta describes self-hosting with Python, FastAPI, PostgreSQL plus pgvector and Redis, and identifies AGPL-3.0 as the license.appsecsanta.com · read 15 Oct 2026
- Integrations
- The maker lists a REST API, Python SDK, Node.js SDK, MCP server, and agent skills for OpenClaw and Claude Code.promptinspector.io · read 4 Oct 2026
- Latency claim
- The maker says most requests finish in the first two layers in under 3 milliseconds and full semantic analysis takes about 30 milliseconds.github.com · read 15 Oct 2026
