robots.txt
See which AI crawlers other sites block →# As a condition of accessing this website, you agree to abide by the following
# content signals:
# (a) If a Content-Signal = yes, you may collect content for the corresponding
# use.
# (b) If a Content-Signal = no, you may not collect content for the
# corresponding use.
# (c) If the website operator does not include a Content-Signal for a
# corresponding use, the website operator neither grants nor restricts
# permission via Content-Signal with respect to the corresponding use.
# The content signals and their meanings are:
# search: building a search index and providing search results (e.g., returning
# hyperlinks and short excerpts from your website's contents). Search does not
# include providing AI-generated search summaries.
# ai-input: inputting content into one or more AI models (e.g., retrieval
# augmented generation, grounding, or other real-time taking of content for
# generative AI search answers).
# ai-train: training or fine-tuning AI models.
# use: how AI systems may consume the content (immediate, reference, or full).
# ANY RESTRICTIONS EXPRESSED VIA CONTENT SIGNALS ARE EXPRESS RESERVATIONS OF
# RIGHTS UNDER ARTICLE 4 OF THE EUROPEAN UNION DIRECTIVE 2019/790 ON COPYRIGHT
# AND RELATED RIGHTS IN THE DIGITAL SINGLE MARKET.
# BEGIN Cloudflare Managed content
# AI crawlers are allowed. This block previously carried `Disallow: /` for
# Amazonbot, Applebot-Extended, Bytespider, CCBot, ClaudeBot,
# CloudflareBrowserRenderingCrawler, Google-Extended, GPTBot and
# meta-externalagent, which shut the site out of exactly the answer engines it
# is trying to be cited in. Deliberately reversed: AI search is this property's
# strongest channel relative to classic organic.
#
# No effect on Google Search either way. Googlebot was never in that list, and
# Google-Extended governs Gemini grounding and training only -- Google documents
# that it does not affect Search ranking or inclusion, and AI Overviews are built
# from the Search index that Googlebot feeds. So this changes nothing about how
# the site ranks; it changes who may read it.
#
# The Content-Signal below is unchanged and still says ai-train=no. Access and
# training rights are separate: crawlers may read and cite this content, and the
# reservation of training rights stands. Content-Signal is a declared restriction
# under Article 4 of EU Directive 2019/790, not an enforcement mechanism.
User-agent: *
Content-Signal: search=yes,ai-train=no,use=reference
Allow: /
# END Cloudflare Managed Content
# A second bare `User-agent: *` group used to sit here carrying only `Allow: /`.
# Two groups for the same agent is ambiguous -- a crawler that takes the first
# match rather than merging would read the signal, one that takes the last would
# not -- and it granted nothing the group above does not already grant. Removed
# so there is exactly one `*` group and it carries the Content-Signal.
# Markdown twins (.md) of every page are published for AI answer engines --
# roughly 87x smaller than the HTML and far cheaper to parse. Search engines
# are told NOT to index them: they duplicate each page's content at a second
# URL, which would split signals for no gain. AI crawlers are welcome to them.
# Same split resources.getbluejay.ai already uses.
#
# The Content-Signal and Allow lines are repeated here on purpose: a crawler
# obeys only its most specific matching group, so once Googlebot has a group of
# its own it stops reading the `User-agent: *` group entirely and would
# otherwise lose the site's content signal.
User-agent: Googlebot
User-agent: Googlebot-Image
User-agent: Bingbot
Content-Signal: search=yes,ai-train=no,use=reference
Allow: /
Disallow: /*.md$
Sitemap: https://getbluejay.ai/sitemap.xml
llms.txt
Compare with other sites' llms.txt files →# Bluejay
> Bluejay is a QA and observability platform for voice and chat AI agents.
> Teams use it to simulate real conversations before launch, load-test at scale,
> monitor live production calls, and run automated evaluations to catch
> regressions and improve agent quality over time.
Bluejay helps engineering and product teams building voice and chat AI agents
test against real-world conversation scenarios — including edge cases,
interruptions, adversarial inputs, and high call volume — replay and monitor
production traffic, and run fine-tuned evaluations and A/B tests across agent
versions and prompts.
## For agents
Every page on this site is available as clean Markdown. Request it with a
`Accept: text/markdown` header and you will get the page body with YAML
frontmatter (title, description, image) instead of rendered HTML — roughly 85x
fewer tokens than the HTML equivalent.
## Product
- [Platform overview](https://getbluejay.ai/platform): Simulation testing across
voice, chat and text; production monitoring; fine-tuned evaluations; A/B
testing for conversational AI agents.
- [Pricing](https://getbluejay.ai/pricing): Pay-as-you-go ($0/mo plus usage, $25
in starting credits, up to 25 concurrent simulations), Growth, Scale and
Enterprise plans. Unlimited seats and agents on every plan. 70+ built-in
metrics plus unlimited custom metrics. SOC 2 Type II; BAA/DPA and SSO/SAML
available.
## Comparisons
- [Compare Bluejay](https://getbluejay.ai/compare): Head-to-head comparisons
against Hamming, Cekura, Coval, Braintrust, Cyara, and Roark, evaluated on
published pricing and capacity, testing and monitoring depth, security
coverage, and closed-loop improvement. Every claim is sourced inline, either
to Bluejay's own pricing and documentation or to the competitor's own public
pages.
- [Bluejay vs. Hamming](https://getbluejay.ai/compare/bluejay-vs-hamming): A
documented Self-Improve workflow and published pricing on every plan,
versus Hamming's "Contact us" tiers.
- [Bluejay vs. Cekura](https://getbluejay.ai/compare/bluejay-vs-cekura):
Included concurrency, speech-quality metrics, and security red teaming.
- [Bluejay vs. Coval](https://getbluejay.ai/compare/bluejay-vs-coval):
Included concurrency and human-calibrated evaluation via Metrics Lab.
- [Bluejay vs. Roark](https://getbluejay.ai/compare/bluejay-vs-roark):
Published usage-based costs and concurrency across plans.
- [Bluejay vs. Cyara](https://getbluejay.ai/compare/bluejay-vs-cyara):
Self-serve pricing with red teaming included, against Cyara's Botium and
Voice Assure product lineup.
- [Bluejay vs. Braintrust](https://getbluejay.ai/compare/bluejay-vs-braintrust):
Built-in caller simulation, a different layer of the stack from a
general-purpose LLM eval platform.
## Answers to common questions
Direct answers on testing and monitoring voice AI agents in production:
- [Voice AI testing answers](https://getbluejay.ai/post): Load testing for
concurrency, tracking task-completion and success rates in production,
running hundreds of automated test calls before release, testing across
languages and accents, and simulating difficult audio conditions.
## Encyclopedia (glossary)
- [Encyclopedia Agentica](https://getbluejay.ai/encyclopedia): 230 defined terms
covering voice and chat AI testing, observability, evaluation, metrics,
compliance, speech technology, telephony infrastructure and the platform
ecosystem — for example adversarial testing, barge-in, word error rate,
webhook, hallucination, agent assist. Each entry defines the term and explains
why it matters when building or testing conversational AI agents.
## Guides and comparisons
- [Guides](https://getbluejay.ai/resources): In-depth guides on testing,
monitoring and evaluating voice and chat AI agents, including enterprise-scale
QA, lean testing for startups, and running evals in a CI/CD pipeline.
- Head-to-head platform comparisons covering Bluejay against Cyara, Cekura,
Coval, Hamming, Roark, Arize, Braintrust, Langfuse and LangSmith, plus
category comparisons (IVR testing vs voice agent testing, voice agent QA vs
software testing, voice AI vs chatbots). All listed on the
[Guides index](https://getbluejay.ai/resources).
## Blog
- [Blog](https://getbluejay.ai/blog): Engineering analysis and the Skywatch
podcast — interviews with founders and engineering leaders shipping production
voice agents, including Presto, Pipecat, Assembled, Telnyx, Deepgram, LiveKit,
Rime and Inworld.
## Company
- [About Bluejay](https://getbluejay.ai/company)
- [Security and compliance](https://getbluejay.ai/security): SOC 2 Type II,
HIPAA with BAAs, and a completed GDPR compliance review. States the nine
security commitments in full -- TLS 1.2+ in transit and AES-256 at rest,
encrypted daily backups, SSO and MFA, least privilege reviewed quarterly,
peer-reviewed pull requests, annual external penetration testing, production
monitoring, a defined incident escalation path with 24-hour customer
notification, and an AI governance commitment that customer data is never
used to train or fine-tune models with AI providers configured for zero data
retention. Lists all fifteen policies and summarises the 36 published
subprocessors with processing regions.
- [Bug Bounty](https://getbluejay.ai/bounty): A closed $6,000 contest, kept up
as a record of the challenge. Three Bluejay voice agents answered real phone
numbers, each with three planted bugs, found using Bluejay simulations.
[Official rules](https://getbluejay.ai/bounty/rules) cover what was in scope.
## Optional
- [Terms of Service](https://getbluejay.ai/terms-of-service)
- [Privacy Policy](https://getbluejay.ai/privacy-policy)
