AI Dispatch
agent · web ai agents

Gaslighting Check vs Harpa AI

55 fields both were evaluated on, 29 of them different — including where one discloses something the other does not. Every value links to the document it came from. 26 further fields are disclosed by neither and are listed at the end rather than tabled.

Assurance

FieldGaslighting CheckHarpa AI
Audit trail·
no public information
harpa.ai · checked Aug 2, 2026
Explainability·
The check points to exact words and common tactics like denial, blame, and DARVO.
www.gaslightingcheck.com · checked Aug 1, 2026
no public information
harpa.ai · checked Aug 2, 2026
Compliance certifications·
Regulatory alignment·
Dispute process·
Purchases are final and not refundable. Please use your free analysis before you upgrade. The only exception is a technical problem that fully blocks the service, like a system error, account access issue, or service outage. Report it within 48 hours to support@gaslightingcheck.com . We review those cases one by one.
www.gaslightingcheck.com · checked Aug 1, 2026
no public information
harpa.ai · checked Aug 2, 2026

Agency

FieldGaslighting CheckHarpa AI
Action space·
read, understand and act upon web page content. It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat.
HARPA can read, understand and act upon web page content. It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat.
harpa.ai · checked Aug 2, 2026
Operating environment·
browser extension, powered by hybrid AI-engine that combines ChatGPT, Gemini, ClaudeAI, Perplexity, DeepSeek with Web Automation.
HARPA AI is a browser extension, powered by hybrid AI-engine that combines ChatGPT, Gemini, ClaudeAI, Perplexity, DeepSeek with Web Automation.
harpa.ai · checked Aug 2, 2026

Safety

FieldGaslighting CheckHarpa AI
Safety policy·
Privacy & safety Built for private conversations The tool checks language patterns. It does not label or diagnose people.
www.gaslightingcheck.com · checked Aug 1, 2026
no public information
harpa.ai · checked Aug 2, 2026
Usage restrictions·
no public information
harpa.ai · checked Aug 2, 2026
Incident reporting·
Email support@gaslightingcheck.com . We will delete your account and related data when you ask.
www.gaslightingcheck.com · checked Aug 1, 2026
no public information
harpa.ai · checked Aug 2, 2026
Data handling·
D Delete anytime Remove your data when you choose.
www.gaslightingcheck.com · checked Aug 1, 2026
does not store your data, and is trusted by 500.000+ professionals and teams of all sizes, with an average review score of 4.7
harpa.ai · checked Aug 2, 2026

Practicality

FieldGaslighting CheckHarpa AI
Pricing model·
Each plan gives you tokens each month. One analysis costs 10 tokens.
www.gaslightingcheck.com · checked Aug 1, 2026
Basic S plan starts with $12 per month paid annually
harpa.ai · checked Aug 2, 2026
Price point·
Basic S plan starts with $12 per month paid annually
harpa.ai · checked Aug 2, 2026
Availability·
HARPA has a basic free tier and a demo period for it's premium features.
harpa.ai · checked Aug 2, 2026
Deployment options·
browser extension
HARPA AI is a browser extension
harpa.ai · checked Aug 2, 2026
Supported regions·
Support model·
Email support@gaslightingcheck.com . We will delete your account and related data when you ask.
www.gaslightingcheck.com · checked Aug 1, 2026
no public information
harpa.ai · checked Aug 2, 2026

Foundation models

FieldGaslighting CheckHarpa AI
Base models·
HARPA supports local models like Meta Llama, does not log your conversations and actions, is GDPR compliant, and can be paired with any LLM model of your choice. HARPA has a hybrid Browser Automation/AI engine and is compatible with OpenAI GPT, Anthropic Claude, Google Gemini, DeepSeek, Perplexity and Meta Llama.
harpa.ai · checked Aug 2, 2026
Model provider·
OpenAI GPT, Anthropic Claude, Google Gemini, DeepSeek, Perplexity and Meta Llama.
HARPA has a hybrid Browser Automation/AI engine and is compatible with OpenAI GPT, Anthropic Claude, Google Gemini, DeepSeek, Perplexity and Meta Llama.
harpa.ai · checked Aug 2, 2026
Model swappable·
can be paired with any LLM model of your choice
harpa.ai · checked Aug 2, 2026
Open weights·
HARPA supports local models like Meta Llama
harpa.ai · checked Aug 2, 2026
Context window·
10,000 to 15,000 characters
For the best speed, keep each analysis under about 10,000 to 15,000 characters, or about 1,500 to 2,500 words.
www.gaslightingcheck.com · checked Aug 1, 2026
8192 tokens on ChatGPT 3.5.
Large Language Models have a limited token window, e.g. 8192 tokens on ChatGPT 3.5.
harpa.ai · checked Aug 2, 2026

Ecosystem

FieldGaslighting CheckHarpa AI
Tool use·
Sort facts, pause before replying, write a safer response, or save a record.
www.gaslightingcheck.com · checked Aug 1, 2026
It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat.
harpa.ai · checked Aug 2, 2026
API access·
HARPA S plan comes with CloudAI API
harpa.ai · checked Aug 2, 2026
Marketplace presence·
official Chrome Web Store.
We recommend installing HARPA from the official Chrome Web Store.
harpa.ai · checked Aug 2, 2026

Impact

FieldGaslighting CheckHarpa AI
User base·
5,000+
Conversations checked 5,000+
www.gaslightingcheck.com · checked Aug 1, 2026
Trusted by 500.000 professionals
harpa.ai · checked Aug 2, 2026
Deployment scale·
HARPA has a basic free tier and a demo period for it's premium features.
harpa.ai · checked Aug 2, 2026
Target sectors·
For people dealing with emotional abuse, coercive control, or conversations that keep getting rewritten.
www.gaslightingcheck.com · checked Aug 1, 2026
no public information
harpa.ai · checked Aug 2, 2026
High-risk domains·
For people dealing with emotional abuse, coercive control, or conversations that keep getting rewritten.
www.gaslightingcheck.com · checked Aug 1, 2026
review CVs, job offers and legal documents
harpa.ai · checked Aug 2, 2026

Disclosed by neither

Both Gaslighting Check and Harpa AI publish nothing on these 26 fields. That is a finding about the category rather than a difference between them, so it is recorded here instead of as 26 identical table rows. Each is shown with its source check on the individual profiles.

  • Autonomy level
  • Human oversight
  • Goal complexity
  • Initiative
  • Documented incidents
  • Integrations
  • Fine-tuning
  • Protocols supported
  • Multi-agent
  • Open source
  • Safety evaluations
  • Red teaming
  • Model or system card
  • Third-party evaluations
  • Insurance available
  • Insurance carriers
  • Coverage limits
  • Indemnification
  • Liability cap
  • Tamper-evident log
  • Runtime governance
  • Permission scopes
  • Outcome-based pricing
  • Settlement mechanism
  • SLA terms
  • Evaluation coverage

Questions

How do Gaslighting Check and Harpa AI compare on action space?
Gaslighting Check: no public information disclosed. Harpa AI: read, understand and act upon web page content. It can navigate, click and extract data from web pages, run it through Large Language Models, paste back to websites or trigger IFTTT chains such as Zapier and Integromat..
How do Gaslighting Check and Harpa AI compare on operating environment?
Gaslighting Check: no public information disclosed. Harpa AI: browser extension, powered by hybrid AI-engine that combines ChatGPT, Gemini, ClaudeAI, Perplexity, DeepSeek with Web Automation..
How do Gaslighting Check and Harpa AI compare on deployment options?
Gaslighting Check: no public information disclosed. Harpa AI: browser extension.
How do Gaslighting Check and Harpa AI compare on model provider?
Gaslighting Check: no public information disclosed. Harpa AI: OpenAI GPT, Anthropic Claude, Google Gemini, DeepSeek, Perplexity and Meta Llama..
How do Gaslighting Check and Harpa AI compare on context window?
Gaslighting Check: 10,000 to 15,000 characters. Harpa AI: 8192 tokens on ChatGPT 3.5..
How do Gaslighting Check and Harpa AI compare on marketplace presence?
Gaslighting Check: no public information disclosed. Harpa AI: official Chrome Web Store..