PromptRedTeam
promptredteam.com
- Free plan
- Web
- API
- Self-hosted
What PromptRedTeam is
PromptRedTeam scans user input for threats before it reaches a large language model. Its detection covers direct prompt injection, role manipulation, zero-width injection, delimiter injection, and encoded payloads. For direct injection it checks for override keywords, instruction conflicts, and prompt structure; role checks cover known jailbreak patterns, role-play scenarios, and bypass attempts. It also detects zero-width characters, normalizes Unicode, analyzes Base64 and other encodings, and can decode suspicious text before scanning it again. Delimiter checks target context-breaking markers, Markdown abuse, and structural manipulation. Applications can send text to the hosted API; the site demonstrates a POST request to https://api.promptredteam.com/test. The live tester accepts up to 1,000 characters and allows 10 requests per minute. The project can also be self-hosted for unlimited usage. Python, JavaScript, and cURL examples are provided. The suggested workflow is to scan input before it reaches an LLM, then block threats or log them for review. PromptRedTeam is open source under the MIT License.
Who it suits
PromptRedTeam suits developers integrating input checks into applications that send text to large language models, or teams that want to self-host the project. Its examples cover Python, JavaScript, and cURL.
Good to know, and what to watch
- Detects five listed categories of prompt attacks
- Hosted API and self-hosted deployment options
- Self-hosting provides unlimited usage
- Implementation examples include Python, JavaScript, and cURL
- Open source under the MIT License
- Live tester allows 10 requests per minute
- Live tester input is limited to 1,000 characters
Plans
Every plan and priceOpen source
Open source
The verdict
PromptRedTeam offers several detection checks and a hosted API or self-hosted option. The live tester has request and input limits; self-hosting is listed as unlimited usage.
What the maker publishes
- Attack coverage
- It detects direct injection, role manipulation, zero-width injection, delimiter injection, and encoded payloads. source
- Company details
- The opened maker page does not state the maker's name, headquarters, or founding year. source
- Delimiter detection
- The site says it detects context-breaking delimiters, markdown abuse, and structural manipulation. source
- Detection categories
- It detects direct injection, role manipulation, zero-width injection, delimiter injection, and encoded payloads. source
- Direct injection detection
- The site says direct injection detection uses pattern matching for override keywords, instruction conflicts, and prompt structure analysis. source
- Hidden and encoded text
- The site describes zero-width character detection, Unicode normalization, Base64 and encoding analysis, and automatic decoding and rescanning. source
- Hosted API
- The homepage demonstrates scanning text with a POST request to https://api.promptredteam.com/test. source
- Hosted API
- The site demonstrates sending text to https://api.promptredteam.com/test using a POST request with JSON. source
- Integration examples
- The homepage provides Python, JavaScript, and cURL implementation options. source
- Integration examples
- The site provides implementation examples for Python, JavaScript, and cURL. source
- License
- The homepage describes PromptRedTeam as open source under the MIT License. source
- Live testing limit
- The live demo allows 10 requests per minute and accepts up to 1,000 characters in its input field. source
- Open source license
- The site describes the project as open source under the MIT License. source
- Purpose
- PromptRedTeam scans user inputs for hidden payloads, role manipulation, and instruction overrides before they reach an LLM. source
- Purpose
- PromptRedTeam scans user inputs for prompt injection before they reach an LLM. source
- Rate limit
- The live tester allows 10 requests per minute and says deploying your own provides unlimited usage. source
- Role manipulation detection
- The site says role manipulation detection recognizes known jailbreak patterns, role-play scenarios, and bypass attempts. source
- Self-hosting
- The homepage says PromptRedTeam can be self-hosted for unlimited usage. source
- Self-hosting
- The site says users can self-host for unlimited usage. source
- Suggested use
- The homepage suggests scanning inputs before they reach an LLM, then blocking threats or logging them for review. source
- Workflow
- The site says applications can scan inputs before they reach an LLM and block threats or log them for review. source
Checked against the maker's pages on 9 Oct 2026.