llmthreat
For vendors 21 vendors · 10 tools Matrix Vendors

Rebuff: Is It Still Maintained in 2026?

Archived

Rebuff is an open-source Guardrails / LLM Firewall tool from Protect AI (a Palo Alto Networks company) — Apache-2.0, written in TypeScript, built for LLM prompt injection detector using heuristics, an LLM-based detector, vector-store similarity, and canary tokens. As of 2026-07-13, its repository status is "Archived". Stars and last-push date are pulled live from the GitHub API.

Repository snapshot

1,511
GitHub stars
2024-08-07
Last push
v0.1.1
Latest release
Apache-2.0
License
TypeScript
Language
Archived
Status

Repo: protectai/rebuff · LLM prompt injection detector using heuristics, an LLM-based detector, vector-store similarity, and canary tokens · pulled live as of , refreshed weekly

What Rebuff does

LLM prompt injection detector using heuristics, an LLM-based detector, vector-store similarity, and canary tokens Protect AI (a Palo Alto Networks company) maintains it under the Apache-2.0 license. Commercial vendors covering the same ground are scored, cell by cell, in the OWASP LLM Top 10 coverage matrix.

Status note: Rebuff is flagged "Archived". Check the repo before you build a pipeline on it. (Confirmed via GitHub's own archive banner: 'This repository was archived by the owner on May 16, 2025.' Matches brief's claim exactly. Last push was 2024-08-07, well before archival.)

Install & source

Repository: https://github.com/protectai/rebuff

Frequently asked questions

Is Rebuff still maintained in 2026?

As of 2026-07-12, Rebuff's repository is flagged "Archived". Check the latest push date on GitHub before you rely on it.

What license is Rebuff under?

Rebuff runs under the Apache-2.0 license, maintained by Protect AI (a Palo Alto Networks company).

What is Rebuff used for?

LLM prompt injection detector using heuristics, an LLM-based detector, vector-store similarity, and canary tokens. It sits in the Guardrails / LLM Firewall category.

Last verified . Sources: https://github.com/protectai/rebuff, Rebuff project documentation. Funding and repo figures are third-party and go stale. Re-verify before you rely on them.
Informational, not professional security advice — verify claims against the maintainer and GitHub before relying on them. Protect AI (a Palo Alto Networks company) and GitHub do not endorse llmthreat.com. OWASP, MITRE, and the listed vendors and maintainers do not endorse llmthreat.com.