Add fak (Fused Agent Kernel) to Agent Firewalls & Gateways#58
Open
anthony-chaudhary wants to merge 1 commit into
Open
Add fak (Fused Agent Kernel) to Agent Firewalls & Gateways#58anthony-chaudhary wants to merge 1 commit into
anthony-chaudhary wants to merge 1 commit into
Conversation
fak is an open-source (Apache-2.0) Go agent kernel that sits on the tool-call path as a default-deny capability gate plus a result quarantine — a direct fit for the Runtime Protection section. Format matches the section's existing entries (bold link, description ending with a period). Signed-off-by: anthony-chaudhary <18080164+anthony-chaudhary@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds fak to the Agent Firewalls & Gateways (Runtime Protection) section.
fak is an open-source (Apache-2.0) Go "agent kernel": a single binary that sits between an AI agent and its tools and adjudicates every tool call on the same call path before it runs — a default-deny capability gate the model can't talk past (no IPC, fails closed), plus a quarantine that holds suspicious tool results out of the model's context to contain prompt injection and tool poisoning. That's a direct match for this section's framing ("tools that sit between the agent and the world to filter traffic, prevent unauthorized tool access, and block prompt injections").
Checklist against the contribution guidelines:
In the interest of honesty (which the project takes seriously — it ships a machine-checked claims ledger), the result detector is evadable by design; the security floor is the capability lock plus the containment, which an attacker has to beat as two independent gates rather than fooling one classifier.