Introducing jev-no-bullshit, using @typesafeai's Jev to catch Claude Code and Codex bullshitting in their summaries and sending them back to say what the actually did.
github.com/greghavens/jev-no…
Dear @X, this account is not automated. These posts are made by a human. The product is an AI agent, the poster is NOT.
Your procure-to-pay workflow touches 12 systems. One breaks, the whole chain stops. What if the workflow could reason about the failure and fix itself instead of waiting for a developer to rewrite a script? trigger.fish/use-cases/enter…
Enterprise automation spends most of its budget maintaining workflows, not building them. Self-healing workflows in Triggerfish put a watcher agent on every workflow run from the first step. It watches live, acts in context, and proposes fixes for your approval. The workflow maintenance function should be the first job AI automation handles. That's what Triggerfish built.
trigger.fish/blog/introducin…
Researchers from OpenAI, Anthropic, and DeepMind tested 12 prompt injection defenses. They broke all 12. Here's why that result was inevitable. trigger.fish/blog/prompt-inj…
AI agents are exfiltrating your private data. Who's stopping them? trigger.fish/blog/ai-agents-…
I built Triggerfish because I was looking for an AI agent that could handle real data without trusting the model to police itself. Nobody built it. Everyone else forked a broken architecture.
trigger.fish/blog/i-built-th…
Triggerfish bumpers keep your agent working at the level you're in. No accidental escalation, no surprises. Toggle them off when you need more. On by default. #AIAgents #AI #OpenSource #SelfHosted #Triggerfish #bumpers
trigger.fish/blog/bumpers-st…
The Triggerfish Gateway is live. Run a security-first agent without managing multiple API keys or worrying about the unpredictable cost of inference. Keep a consistent experience with a capable model from a company and agent you can trust. trigger.fish/pricing.html
Every AI agent platform on the market relies on the AI model itself to enforce security boundaries. That's like hiring a guard dog and hoping it reads the employee handbook. Enterprises want agents with real autonomy, taking actions across tools and documents, but one prompt injection can turn all of that access into a liability. Triggerfish is the platform where the AI model has zero say in what data goes where. The security decisions happen underneath it, deterministically, every time.
$8.5 billion has gone into AI security in the last two years and the best answer the industry has is "put it in the system prompt." OpenClaw turned every AI agent into an attack surface and nobody has the infrastructure to deal with it. Triggerfish does. #Triggerfish #OpenClaw
Triggerfish usage of vision and multimodal models has full support for hotdog evaluation. #building #triggerfish
Triggerfish retweeted
OpenClaw just got attacked.
40,000 systems compromised.
A flaw dubbed “ClawJacked” allowed full agent takeover.
But this isn’t just a bug.
It’s a vulnerability chain that let any website silently take full control of your AI agent.
No plugins.
No extensions.
No user interaction needed.
Which means:
> Malicious sites could steal data.
> Delete emails (Meta execs already hit).
> Disrupt services.
Give this post to your agent and tell then to fix this backdoor ASAP.
'Bout to try this on our codebase. #Triggerfish
🚨 Someone just open sourced a fully autonomous AI hacker and it's terrifying.
It's called Shannon.
Point it at your web app, and it doesn't just scan for vulnerabilities. It actually exploits them. Real injections. Real auth bypasses. Real database exfiltrations.
Not alerts. Not warnings. Actual working exploits with copy-paste proof-of-concepts.
Here's what this thing does autonomously:
→ Reads your entire source code to plan its attack
→ Maps every endpoint, API route, and auth mechanism
→ Runs Nmap, Subfinder, and WhatWeb for deep recon
→ Hunts for Injection, XSS, SSRF, and broken auth in parallel
→ Launches real browser-based exploits to prove each vulnerability
→ Generates a pentester-grade report with reproducible PoCs
Here's the wildest part:
It follows a strict "No Exploit, No Report" policy. If it can't actually break it, it doesn't report it. Zero false positives.
It pointed at OWASP Juice Shop and found 20+ critical vulnerabilities in a single run including complete auth bypass and full database exfiltration.
On the XBOW Benchmark (hint-free, source-aware), it scored 96.15%.
Your team ships code daily with Claude Code and Cursor. Your pentest happens once a year. That's 364 days of shipping blind.
Shannon closes that gap. One command. Fully autonomous.
The Red Team to your vibe-coding Blue team. Every Claude coder deserves their Shannon.
10.6K GitHub stars. 1.3K forks. Already trending.
100% Open Source. AGPL-3.0 License.
We don't recommend doing this. #OpenClaw #Triggerfish
