Prediction Didn’t hold up AI assessment confidence: 92% certainty 5/5 debate potential 2/5

Finn: Anti-prompt-injection skills will launch within two weeks

Alex Finn · Clawdbot/OpenClaw Clearly Explained (and how to use it) · Jan 27, 2026 · at 32:05

Alex Finn outlines the emerging security mitigations for AI agents interacting with incoming public messages.

0:00 / 0:10exact quote · 10.4s

▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark

“I'm sure I'm very positive in the next week or two, there'll be official skills out there that like handle prompt injection from email, prompt injection from tweets and replies.”

quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →

The claim, plainly

Official skills handling prompt injection from emails and social replies will be released for Claudebot within one to two weeks.

a quote alone can be ambiguous, so this restates it with the references and context of the conversation around it resolved. AI-written; the tape below the player is the source.

Assessment

Alex Finn predicted on January 27, 2026, that official skills explicitly mitigating prompt injection from emails and social media replies/tweets would be released for the project (Clawdbot/Claudebot, subsequently rebranded to Moltbot and OpenClaw) within one to two weeks (by February 10, 2026). As of the February 10, 2026 deadline, maintainer Peter Steinberger had published no dedicated prompt-injection defense skills on the official ClawHub directory. Instead, repository issue discussions in early February 2026 documented the maintainer's position that dedicated prompt-injection scanning was left to third-party community plugins (such as ClawGuard and ClawMoat) rather than native official skills.

Peter Steinberger's official ClawHub skills catalog did not include dedicated skills for prompt injection filtering for email or social media replies as of February 10, 2026.
OpenClaw repository issue #7705 on February 2, 2026, confirmed that dedicated prompt-injection scanning tooling was delegated to third-party community plugins rather than core official skills.
Core security updates during early February 2026 implemented general system hardening and delimiter handling, but no official specialized prompt-injection defense skill was released within the stated window.

Checked against

[1] snyk.io [2] wikipedia.org [3] github.com [4] github.com [5] websearchapi.ai [6] github.com [7] clawbot.ai [8] github.com

Peter Steinberger Claudebot skills prompt injection release February 2026 "ClawHub" "prompt injection" skill steipete steipete clawdbot OR claudebot prompt injection email social skills github steipete claudebot prompt injection skills "prompt injection" "steipete" site:github.com "prompt injection" "email" "steipete" site:github.com/openclaw "prompt injection" "skill" "steipete" site:github.com

assessed as of 2026-09-02

This assessment is an AI reading of a public statement, made against the cited references with the confidence shown. It can be wrong. The quote, the tape and the references are all on this page; if something is off, use report an error below. full disclaimer →

Mentioned

Alex Finn Claudebot

⚑ Spot an error? Report it

More from Alex Finn

Opinion

Finn: A computer science degree is the most useless degree today

“I have a computer science degree, which is probably the most useless degree you can possibly get at this point in time.”

Alex Finn Feb 25, 2025 ▶ 57:33 VIBE CODING A COOL VIDEO GAME (Grok 3 + Claude 3.7)

Prediction Not checkable as stated

Finn: Hardware bottlenecks and soaring memory costs will persist long-term

“My, kind of thesis, which I think is playing out in front of us right now is we're going to have hardware bottlenecks for a very long time now because the AI build out's happening. And so memory sky, I just bought a memory card for my Nintendo switch to cost m…”

Alex Finn Jun 6, 2026 ▶ 39:18 Hermes Agent Desktop: Full Setup + Real Use Cases

Prediction Not checkable as stated

Finn: Local multi-agent pipelines will automate video production in 45 seconds

“Now five different models locally on my computer are going to basically process this video and in 45 seconds The entire production process is over, and it's going to be uploaded on YouTube, right? That's where this is going.”

Alex Finn Jan 27, 2026 ▶ 21:26 Clawdbot/OpenClaw Clearly Explained (and how to use it)

Prediction Not checkable as stated

Finn: Lead gen returns for interactive AI apps will vanish in three years

“Like we're in that beginning moments of this strategy where people who attack it and use it are going to see unbelievable returns. And then in three years when everyone's doing it, the returns won't be there.”

Alex Finn Feb 25, 2025 ▶ 1:02:53 VIBE CODING A COOL VIDEO GAME (Grok 3 + Claude 3.7)

Opinion

Alex Finn: Hermes Overtook OpenClaw with Desktop Release

“I believe this is like the moment that Hermes overtook OpenClaw, to be quite honest”

Alex Finn Jun 6, 2026 ▶ 2:09 Hermes Agent Desktop: Full Setup + Real Use Cases

Opinion

Finn: Claude Opus 4.8 is the smartest AI model available

“I think Opus for eight is the smartest AI model out there.”

Alex Finn Jun 6, 2026 ▶ 8:05 Hermes Agent Desktop: Full Setup + Real Use Cases

Finn: Anti-prompt-injection skills will launch within two weeks

More from Alex Finn

Turn any episode into a week of clips.