TLP 不是建議：為何協調披露必不可少

TLP Is Not a Suggestion | Akamai

[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion)[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion)[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion)[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion)

![Image 4: Akamai](blob:http://localhost/a2aaba961e4b54e60dc2c572c6a103d8)

[](javascript:void(0);)

English

Back Language Close
* [English](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion "English")
* [Deutsch](https://www.akamai.com/de "Deutsch")
* [Español](https://www.akamai.com/es "Español")
* [Français](https://www.akamai.com/fr "Français")
* [Italiano](https://www.akamai.com/it "Italiano")
* [Português](https://www.akamai.com/pt "Português")
* [中文](https://www.akamai.com/zh "中文")
* [日本語](https://www.akamai.com/ja "日本語")
* [한국어](https://www.akamai.com/ko "한국어")

Back Log in Close

Cloud Manager Manage your cloud computing services

Back Log in Close

Control Center Manage your security and delivery services

* Products

Back Products Close

Back
1. Cloud Computing
2. Cybersecurity
3. Content Delivery
4. See all products
5. Our Infrastructure
6. Global Services

Back

Back Cloud Computing Close

Artificial intelligence (AI)
* Akamai Inference

Databases
* Managed Databases

View cloud pricing

Explore plans and pricing that fit your needs — from small projects to global-scale deployments.

See pricing

Get started with Akamai Cloud

See all Cloud Computing

Back Cybersecurity Close

Segmentation
* Akamai Guardicore Segmentation

zero trust security
* Akamai Workforce Protector (formerly LayerX)
* Secure Internet Access
* Enterprise Application Access
* Akamai MFA
* Identity, Credential, and Access Management

Get started with Security

Protect the applications that drive your business — every day, every time.

Contact Sales

See all Cybersecurity

Back Content Delivery Close

Application performance
* Ion
* API Acceleration
* IP Accelerator

Monitoring, reporting and testing
* Data Stream
* mPulse
* CloudTest

Get started with Content Delivery

Trust the agility and scale of Akamai to help you flawlessly deliver extraordinary digital experiences.

Contact Sales

See all Content Delivery

Solutions

Back Solutions Close

Pricing

Back Pricing Close

Security and Delivery

Get started
* Contact Sales
* Free trials

Cloud pricing

PRODUCT-SPECIFIC PRICING
* GPU
* Kubernetes
* CPU
* Object Storage
* See all pricing

Cloud pricing

Try Akamai Cloud with US$100 in credits*

Deploy faster with global cloud infrastructure — no surprise bills, no lock-in, and transparent pricing across every data center.

Try now

*SeePromotion Redemption Rules & Conditions

Developers

Back Developers Close

Cloud developers
* Developer hub
* Akamai GitHub repo

docs and guides
* Cloud docs
* Guides and tutorials

cloud marketplace
* Developer apps

Get started with Akamai Cloud

Resources

Back Resources Close

Learning
* White papers, ebooks, videos, product briefs
* Customer stories
* Training and certifications

Cybersecurity Research
* Akamai Security Intelligence Group (SIG)
* State of Internet (SOTI) reports

[](https://www.akamai.com/channel-partners)

Partners

Partner with Akamai to innovate, scale, and grow your advantage

Log inBack Log in Close Cloud Manager Manage your cloud computing services Back Log in Close Control Center Manage your security and delivery services
* Docs
* Partners
* Sales
* Support
* Under Attack ?

English

Create account

Under Attack?

Back Create Account Close

##### Akamai Cloud

##### Akamai Security and Delivery

Connect with our Sales team to discuss your business needs and find the right solutions.

Contact Sales

Image 6: Background

1. Security Research
2. /
3. TLP Is Not a Suggestion

TLP Is Not a Suggestion

October 22, 2026 by Akamai Security Intelligence Group

TLP Is Not a Suggestion

1. What TLP is for
2. What breaking the discipline has cost
3. What respecting the process looks like
4. The thing that is actually at stake
5. Sources
1. PHP2Shell, October 2026
2. Exposure counts
3. TLP standard
4. Documented cases

[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion "Share on Twitter")[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion "Share on Facebook")[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion "Share on LinkedIn")[](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion "Share via Email")

On October 21, 2026, the PHP project shipped an emergency fix for PHP2Shell, two critical remote code execution bugs. The release was scheduled for Thursday, October 22. It went out on Wednesday because a Hacker News post broke the embargo on the previous Monday, a vendor publicly confirmed the bugs within hours, a researcher published a reconstruction before the patch existed, and exploitation attempts started before most administrators knew there was anything to patch.

The behavior that followed it is the one we want to talk about: the scramble among people with privileged knowledge to break news first.

Exposure estimates run from roughly two million hosts to more than five million depending on who is counting. The leak itself was one failure. The behavior that followed it is the one we want to talk about: the scramble among people with privileged knowledge to break news first.

None of the scrambling should have been necessary, because the industry has spent 25 years building a process for exactly this scenario. Coordinated disclosure gives a maintainer time to fix a vulnerability before the world learns what to attack. Trust groups and embargo lists give downstream vendors enough notice to have protections ready the moment the patch lands.

The Traffic Light Protocol (TLP), created in 1999, gives everyone in that chain a shared vocabulary for who may know what, and when. The process is not perfect and it is not legally binding. It works because the participants want it to, and because each of them accepts that they will sometimes know something they cannot yet say.

To be precise, the vendor that confirmed the leak withheld technical details until the patch shipped, and that is the right call. The independent researcher who published a proof of concept before the release did _not_. The distinction matters, and the industry has a framework for exactly this scenario, and it is worth respecting even when it costs you a news cycle. Restraint to align on what is shared, when it is communicated, and how remediations are recommended is a foundational principle that threat fighters adhere to for the public good.

That restraint is the subject of this blog post. Vulnerability response is a team sport. The maintainers, the researchers, the platform vendors, the hosting providers, and the security vendors are competitors in most contexts — but, inside an embargo, they are all on one side and the attacker is on the other.

The payoff is a market that is safer for everyone than any one participant could make it alone. The moment a participant treats early access as a marketing asset by racing to confirm, to reproduce, or to publish first, the game turns zero sum: That vendor’s brand gains a news cycle and every customer outside the circle pays for it in hours of exposure.

PHP2Shell is the latest example of what that race costs, and of why the protocol deserves respect even when honoring it means being quiet while someone else is being loud.

What TLP is for

The TLP exists so that people who distrust one another commercially can still share what they know about an active threat. It was created in 1999 by the United Kingdom's National Infrastructure Security Co-ordination Centre, standardized by the Forum of Incident Response and Security Team (FIRST) in 2015, and revised to version 2.0 in August 2022 after a working group of more than 50 incident responders rewrote the definitions for clarity (FIRST release,CERT-EU primer).

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) adopted 2.0 for US government sharing, the Information Sharing and Analysis Centers (ISACs) adopted it for sector sharing, and FIRST membership is conditional on using it.

The TLP’s four labels are simple, and the simplicity is the point (Table 1).

Label	Who may receive the information	What it means in practice
TLP:RED	The named recipients only	Not for disclosure — only discussed in a briefing by the people in the room
TLP:AMBER+STRICT	The recipient's own organization	No clients, no partners, no subcontractors
TLP:AMBER	The recipient's organization and its clients, on a need-to-know basis	The ordinary embargo label for pre-patch vulnerability detail
TLP:GREEN	The recipient's community	Peers and partner organizations, never public channels
TLP:CLEAR	Anyone	(Formerly TLP:WHITE) Standard copyright applies, nothing else

Table 1: The four labels of the TLP and what they mean

The protocol only works among people who want it to work, and it fails the moment a participant decides their standing in the market matters more than their standing in the community.

The standard says two things that people routinely forget. First, the source sets the label and only the source can change it. A recipient who decides the information is “basically public now” and re-labels it has not downgraded the label. They have broken it.

Second, TLP is a sharing protocol, not a classification scheme or a legal control. It has no enforcement mechanism beyond reputation and continued access. That is a feature and here’s what it means: The protocol only works among people who want it to work, and it fails the moment a participant decides their standing in the market matters more than their standing in the community.

That second point is what the “first to defend posture” gets wrong. The value of being on an embargo list is that you receive information early so that your customers are protected when the patch lands. The value is not that you receive information early so that you can tell the world you received it early. Those are different products. One is what the list was built to produce. The other is what a marketing team extracts from it.

What breaking the discipline has cost

The mechanism for breaking the discipline is the same in each case. Privileged information left the circle it was given to, and the people outside that circle paid for it (Table 2).

Case	Year	What left the circle	Documented consequence
Heartbleed, OpenSSL	2014	Early notice went to a handful of companies chosen by the discoverers rather than to the distributor’s pre-disclosure list; Cloudflare patched and blogged that it had been notified 12 days earlier	Red Hat, Debian, Ubuntu, and most of the hosting industry learned of the bug at public release and spent days exposed while the early recipients advertised that they were already safe. The oss-security thread records the Red Hat response in real time (Openwall). The brand, logo, and domain shipped before the coordinated advisory did.
Meltdown and Spectre	2018	A six-month embargo collapsed six days early after press reporting pieced it together from public kernel commits; Google published early citing growing speculation	Vendors who planned on the agreed-upon date (six days later) were still testing when the details went public (All About Circuits). Cloud providers outside the Tier 1 list found out from the news (LWN panel). Four members of the US House of Representatives wrote to Amazon, AMD, Apple, Arm, Google, Intel, and Microsoft questioning whether embargoes work at all.
Log4Shell, Alibaba Cloud	2021	The reverse case: Alibaba's researcher reported to Apache on November 24 under coordinated disclosure, and details reached GitHub on November 30, before the December 6 patch	China's MIIT suspended Alibaba Cloud from its threat-sharing platform for six months for not reporting to the state first. This shows what happens when a government decides the sharing circle should be itself: The discloser gets punished for doing the right thing, and every Chinese MAPP partner now carries a conflicting legal obligation (The Register).

Table 2: Examples of vulnerability disclosures that violated embargoes, and the resulting impact

A pattern runs through this table: The damage lands outside the circle of trust. In every case, the early recipients were fine. Red Hat's customers and the Tier 2 clouds were not. The embargo survives not because it is enforced but because the people inside it mostly choose to honor it.

Heartbleed deserves one more sentence because it is the closest analog to the current situation. Nobody leaked an exploit. A small group received early warning through informal channels, used it well, and then told everyone that they had used it well — while the rest of the industry was still reading the advisory. That is the first-to-defend posture in its purest form, and the oss-security archive shows exactly how the distributors who carried the actual patching burden felt about it.

What respecting the process looks like

None of this requires silence. It requires sequencing. The seven practices below are not new; they are what the standard already asks for, restated for a marketing calendar.

1. The source sets the label and the clock. If you received it under TLP:AMBER, it stays AMBER until the source says otherwise. A leak elsewhere does not change your label. The maintainer may choose to accelerate; you do not get to choose for them.

2. Protect first, publish second, in that order, and with a gap. Push the signature, the rule, the virtual patch. Tell your customers privately that they are covered. There is a critical dimension that exists, however, with this process. Mitigation partners must still abide by embargo restrictions to not divulge sensitive vulnerability details prior to the public release. In practice, this means that customers must not have access to the raw detection logic. Additionally, any evidence samples that resulted from new detection should be minimal and not reveal data that could be used for full exploit reconstruction (see number 4). The public post waits for the vendor advisory, and it links to the advisory rather than replacing it.

3. Confirm existence only once the source has. Once the maintainer has announced a release, saying that “This is real, patch at 14:00 UTC” is useful amplification. Saying “We reproduced it” before the fix exists tells attackers the leaked details are good enough to work from.

4. No reconstructions, no proofs of concept, no root cause before the fix. This is the line that was crossed with PHP2Shell. It does not matter whether the publisher is a vendor or an individual with a following.

5. No CVE numbers before the CNA publishes them. Circulating identifiers before CVE.org has assigned them puts your guess into other people's advisories. It was wrong at least once for PHP2Shell.

6. When you are first and wrong, say so in the same place you were first. Corrections buried in an update log do not undo the advisories that copied the original.

7. Measure the team on coverage at patch time, not on the timestamp of the blog post. If the KPI is to be first to publish, the behavior follows the KPI. Change the KPI.

And a note for the vendors who do this well, because they exist. Confirming this leak was real, withholding details, and shipping detection guidance alongside the maintainer's advisory is the right shape. The discipline is not about saying nothing. It is about saying nothing that the people writing exploits could not already have worked out for themselves.

The thing that is actually at stake

The TLP has no teeth because it was never meant to need them. It is a promise that competitors make to one another so that the people they both serve are safer than either could make them alone. Every leak in Table 2 was committed by someone who had been trusted with that promise, and every one of them ended with a smaller circle, a slower briefing, or a program that nearly got shut down. The customers who paid were never the ones who broke the protocol.

There will be a postmortem on how PHP2Shell leaked. That is worth doing. The more useful question for the industry is how many of us, upon seeing the leak, reached for a keyboard before reaching for the standard.

If being first to defend means being first to publish, the protocol loses a participant and the next embargo is shorter. If it means being first to protect your customers and last to talk about it, the protocol gains one, and the next embargo holds. Pick the one you would want the other vendors on the list to pick.

Sources

About the author

Image 7: Akamai Wave Blue

Akamai Security Intelligence Group

Akamai SIG is a global team of world-class researchers, engineers, strategists, and data scientists with a broad range of expertise and security disciplines. Our data sources include the enormous Akamai Cloud, open sources, collaboration with third parties, and dark web intelligence. We have also developed our own algorithms and tools that help us deliver our research and keep Akamai security solutions up to date.

View author profile

Related Blogs

Image 8: A vulnerability with no known exploit today may have one tomorrow.

Node2Shell (CVE-2026-111497) is an out-of-bounds write in the Brotli decoder bundled with Node.js that can lead to RCE.

Security Research

Inside Node2Shell: A Closer Look at CVE-2026-111497

October 26, 2026Akamai Security Intelligence Group

Learn about Node2Shell (CVE-2026-111497), a critical Brotli flaw in Node.js enabling RCE. Discover affected versions, detection queries, and mitigation steps.

Image 9: A vulnerability with no known exploit today may have one tomorrow.

The two critical vulnerabilities known as PHP2Shell allow attackers to achieve RCE on servers running PHP 5 and above.

Security Research

Understanding PHP2Shell: Everything You Need to Know

October 22, 2026Akamai Security Intelligence Group

Discover how the critical PHP2Shell vulnerabilities (CVE-2026-109618 & CVE-2026-109619) allow remote code execution in PHP, plus key mitigations.

Image 10: A vulnerability with no known exploit today may have one tomorrow.

A vulnerability with no known exploit today may have one tomorrow.

Research

Beyond Day Zero: Akamai's Rapid Response and AI Threat Intelligence

October 22, 2026Akamai Security Intelligence Group

Discover how Akamai responded to critical PHP vulnerabilities using agentic AI, real-time telemetry, and rapid action switching to deliver adaptive WAF defense.

Products

Company

Careers

Newsroom

Legal & Compliance

Glossary

[](https://twitter.com/Akamai)
[](https://www.facebook.com/AkamaiTechnologies/)
[](https://www.linkedin.com/company/akamai-technologies)
[](https://www.youtube.com/user/akamaitechnologies)

English

Back Language Close

[English](https://www.akamai.com/blog/security-research/tlp-is-not-a-suggestion "English")
[Deutsch](https://www.akamai.com/de "Deutsch")
[Español](https://www.akamai.com/es "Español")
[Français](https://www.akamai.com/fr "Français")
[Italiano](https://www.akamai.com/it "Italiano")
[Português](https://www.akamai.com/pt "Português")
[中文](https://www.akamai.com/zh "中文")
[日本語](https://www.akamai.com/ja "日本語")
[한국어](https://www.akamai.com/ko "한국어")

Your cookie choices and AI settings for this website

We use cookies to ensure the fast reliable and secure operation of this website, to improve your website experience, to enable certain social media interactions, to enable our AI sales representative chatbot (KAI) and to manage your cookie choices. Some cookies process personal data or personal information, where applicable. By continuing to visit our websites you are agreeing to our use of cookies and to the related processing activities. Click “Manage Preferences” to make individual choices and get details on the cookies in use and the associated processing activities in the Cookie Details section. You can access the Cookie Management Page and withdraw the consent at any time via the Cookie Settings link in the footer. For additional information relating to your privacy take a look at our Privacy Statement

Manage Preferences

Privacy Preference Center

When you visit any website, it may store or retrieve information on your browser, mostly in the form of cookies. This information might be about you, your preferences, or your device, and is mostly used to make the site work as you expect. The information does not usually identify you directly, but it can give you a more personalized web experience. Because we respect your right to privacy, you can choose not to allow some types of cookies. Click on the different category headings to learn more and change our default settings. Blocking some types of cookies may impact your experience of the site and the services we are able to offer.

Privacy Statement

Allow All
### Manage Consent Preferences

Performance Cookies

[x] Performance Cookies

These cookies allow us to count visits and traffic sources so we can measure and improve the performance of our site. They help us to know which pages are the most and least popular and see how visitors move around the site. The cookies may be set by us or third party providers. All information these cookies collect is aggregated and therefore anonymous. If you do not allow these cookies we will not know when you have visited our site, and will not be able to monitor its performance.

Cookies Details

Functional Cookies

[x] Functional Cookies

These cookies enable the website to provide enhanced functionality and personalization. They may be set by us or by third party providers whose services we have added to our pages. If you do not allow these cookies then the enhanced functionality and personalization will not be available to you but only the basic functionalities.

Cookies Details

Targeting Cookies

[x] Targeting Cookies

These cookies are set through our site by us or our advertising partners. They may be used by us or those companies to build a profile of your interests and show you relevant adverts on our or other sites. They are based on uniquely identifying your browser and internet device. If you do not allow these cookies, you will experience less targeted advertising, adverts will be shown to you, nevertheless.

Cookies Details

Strictly Necessary Cookies

Always Active

These cookies are necessary for the website to function and cannot be switched off in our systems. They are usually only set in response to actions made by you which amount to a request for services, such as setting your privacy preferences, logging in or filling in forms.

Cookies Details

Social Media Cookies

[x] Social Media Cookies

These cookies are set through our site by a range of third party social media services that we have added to the site to enable you to share our content with your friends and networks. These cookies are capable of tracking your browser across other sites and building up a profile of your interests. This may impact the content and messages you see on other websites you visit. If you do not allow these cookies you may not be able to use or see the social media sharing tools.

Cookies Details

AI ChatBot Cookies

[x] AI ChatBot Cookies

These cookies are necessary for the operation of the AI chat bot on the website. By agreeing to the use of these cookies you are also agreeing on the activation of the AI chatbot powered by Open AI GPT 4 to learn more about our products and offerings. The cookies placed by the AI chat bot are functional cookies meaning necessary for the AI chatbot to operate but not necessary for the operation of the website as such. They assist us sharing the information you are requesting via the chatbot and engaging with you.

Cookies Details

Cookie List

Clear
* - [x] checkbox label label

Apply Cancel

Consent Leg.Interest

[x] checkbox label label

[x] checkbox label label

[x] checkbox label label

Confirm My Choices

TLP 不是建議：為何協調披露必不可少

TLP Is Not a Suggestion | Akamai

View cloud pricing

Get started with Akamai Cloud

Get started with Security

Get started with Content Delivery

Security and Delivery

Cloud pricing

Cloud pricing

Try Akamai Cloud with US$100 in credits*

Get started with Akamai Cloud

Partners

TLP Is Not a Suggestion

TLP Is Not a Suggestion

What TLP is for

What breaking the discipline has cost

What respecting the process looks like

The thing that is actually at stake

Sources

PHP2Shell, October 2026

Exposure counts

TLP standard

Documented cases

About the author

Akamai Security Intelligence Group

Tags

Related Blogs

Inside Node2Shell: A Closer Look at CVE-2026-111497

Understanding PHP2Shell: Everything You Need to Know

Beyond Day Zero: Akamai's Rapid Response and AI Threat Intelligence

Your cookie choices and AI settings for this website

Privacy Preference Center

Performance Cookies

Functional Cookies

Targeting Cookies

Strictly Necessary Cookies

Social Media Cookies

AI ChatBot Cookies

Cookie List