Traffic Light Protocol (TLP) v2.0

SkillDev tools

Use when the user asks "what TLP should this be?", applying TLP markings to a finished product, or the tradecraft pipeline calls for TLP determination before sharing. Covers TLP v2.0 (CLEAR / GREEN / AMBER / AMBER+STRICT / RED).

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Traffic Light Protocol (TLP) v2.0 skill

What this skill tells your AI

The instructions your AI receives, as published by liberty91ltd/cti-skills in skills/tlp-guide/SKILL.md and read by ahel’s review.

Every intelligence output from this platform MUST have a TLP designation. This is enforced by platform rules and validated by hooks.

TLP Definitions

TLP:RED — Named Recipients Only

When to use: Information that would cause significant harm if shared beyond the named recipients. Sources may face risk.

  • Share with: Only the named recipients, in person or via secure channels
  • Do NOT share with: Anyone not explicitly named, even within the same organisation
  • Examples: Active zero-day exploitation details shared to a specific CISO; source-identifying intelligence; ongoing law enforcement operation details

TLP:AMBER+STRICT — Organisation Only

When to use: Information that requires support to be effectively acted upon, limited to the recipient's organisation only.

  • Share with: Members of the recipient's own organisation on a need-to-know basis
  • Do NOT share with: External parties, clients, partners, or affiliated organisations
  • Examples: Targeted threat intelligence specific to one organisation's infrastructure; internal investigation findings

TLP:AMBER — Organisation + Need-to-Know

When to use: Information that requires support to be effectively acted upon, and may be shared with clients/partners on a need-to-know basis.

  • Share with: Members of the recipient's organisation AND their clients/partners who need the information to protect themselves
  • Do NOT share with: The general public, social media, or parties without a clear need-to-know
  • Examples: Sector-specific threat briefings shared with ISAC members; IOCs for active campaigns affecting a supply chain

TLP:GREEN — Community

When to use: Information useful for awareness within a broader community but not for public disclosure.

  • Share with: The wider community (sector peers, security community) but not via publicly accessible channels
  • Do NOT share with: The general public, social media, public blogs, or news outlets
  • Examples: General threat landscape updates shared at closed conferences; new TTP descriptions shared within an ISAC

TLP:CLEAR — Unrestricted

When to use: Information that carries minimal or no foreseeable risk of misuse. Public sharing is acceptable.

  • Share with: Anyone, through any channel
  • Examples: Published CVE details, public vendor advisories, general best practices, published threat reports

Decision Tree

Is this information source-identifying or would sharing risk the source?
  → YES → TLP:RED

Would sharing beyond the recipient's org cause harm?
  → YES → Does it need to stay within a single org?
    → YES → TLP:AMBER+STRICT
    → NO (partners/clients may need it) → TLP:AMBER

Is this useful for community awareness but not public?
  → YES → TLP:GREEN

Is this safe for public consumption?
  → YES → TLP:CLEAR

Default TLP by Product Type

Product TypeDefault TLPRationale
Flash report (active incident)AMBEROften contains org-specific context
Threat actor profileGREENGeneral awareness value
Campaign reportAMBERMay contain victim-identifying information
IOC list (enriched)GREENBroadly useful for detection
Threat assessmentAMBEROften tailored to specific stakeholder
Detection rules (SIGMA/YARA/KQL)GREENBroadly useful for defence
STIX bundleGREENDesigned for automated sharing
Strategic landscape reportGREENGeneral awareness

These are defaults — always evaluate the specific content and adjust.

Implementation

In file frontmatter:

tlp: GREEN

In report headers:

TLP:GREEN — This information may be shared within the security community
but should not be published on publicly accessible channels.

Common Mistakes

  • Over-classifying (TLP:RED for everything) — reduces trust and sharing
  • Under-classifying (TLP:CLEAR for IOCs with victim context) — risks harm
  • Not including TLP at all — violates platform rules
  • Forgetting to downgrade TLP when information becomes public
  • Sharing TLP:AMBER information on public Slack channels or social media

Signals

GitHub stars
22
Forks
9
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
tlp-guide
Source
github.com/liberty91ltd/cti-skills