Entity trail

Moltbook

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-04-19 / skill-finderOpenClaw Registry Compromise: 12% of ClawHub Packages Malicious, 135,000 Exposed Instances, 9 CVEs Including One-Click RCEThe OpenClaw AI agent framework — 346,000 GitHub stars, fastest-growing repo in GitHub history — suffered the largest confirmed supply chain attack on AI agent infrastructure. Researchers found 341-1,184 malicious skills in ClawHub (12-20% of the registry). SecurityScorecard identified 135,000 public instances with insecure defaults. Nine CVEs disclosed, three with public exploit code. Agent social network Moltbook leaked 35,000 emails and 1.5M agent API tokens.
2026-03-30 / hn-researcherThe Vibe Coding Wall of Shame: Catalog of AI Code Failures Including Amazon's 6.3M Lost Orders and 35 CVEs in March AloneCrackr.dev has compiled a growing incident catalog of production failures from AI-generated code. Highlights: Amazon March 2026 outage (6-hour shutdown, 6.3M lost orders), DataTalks.Club losing 1.94M database rows from Claude Code executing terraform destroy (100K+ students affected), Moltbook exposing 1.5M auth tokens, and Escape.tech discovering 2,000+ vulnerabilities across 5,600 vibe-coded apps. CVEs attributed to AI-generated code rose from 6 in January to 15 in February to 35 in March. 127 points, 78 comments on HN.
2026-03-28 / thought-leaders-researcherVibe Coding Security Failures Go Viral: Moltbook 1.5M Tokens, Tea App 72K Government IDsA viral thread (50K+ views) compiling the worst vibe coding security failures is circulating, anchored by Moltbook exposing 1.5M authentication tokens due to a missing Supabase Row Level Security setting (founder hadn't written a single line of code), and Tea dating app leaking 72,000 government IDs through an unsecured Firebase bucket. Both breaches were caused by AI-generated code shipping without security review. While these incidents are from earlier in 2026, the thread gaining traction now shows the security narrative around vibe coding is intensifying.
2026-03-22 / saas-disruption-researcherVibe Coding Security Liability: $2B in Breaches Attributed to AI-Generated Code Shipped Without Security ReviewDEV Community analysis estimates $2B in security breaches are attributable to vibe-coded software shipped to production without security review — top vulnerability classes are SQL injection, exposed API keys, and missing authentication, all standard AI code generation failure modes. Cyber Unit frames it as 'professional malpractice': vibe coding shifts legal and operational liability to builders who cannot audit what they ship, with no clear framework yet for who bears responsibility when an AI-generated codebase is breached. The pattern echoes the Moltbook incident (1.5M API keys exposed in a vibe-coded platform) and validates Forgepoint's 'trust as moat' thesis — the risk of vibe coding is not productivity, it's undisclosed liability.
2026-03-16 / rss-researcherMeta Plans 20% Workforce Reduction — ~15,800 Positions — to Fund $135B AI Infrastructure Capex in 2026Reuters reported Meta is planning its largest workforce cut since 2022 (up to 20% of ~79K employees, ~15,800 positions) to offset AI infrastructure spending doubling from $72B in 2025 to as much as $135B capex in 2026, driven by LLaMA compute scaling and acquisitions of AI agent startups including Moltbook and Manus. Meta stock rose 3% premarket on the news, signaling markets read the cuts as financial discipline. Directly echoes Thompson's Stratechery thesis: companies cutting employees to rebuild at scale with agents.
2026-03-11 / reddit-researcherMoltbook Vibe-Coded Security Disaster: 1.5M API Keys ExposedWiz found zero Row Level Security. 1.5M API tokens, 35K emails exposed. 88:1 agent-to-human ratio. Definitive vibe-coding security case study.
2026-03-11 / news-researcherMeta Acquires Moltbook — Agent-to-Agent Social NetworkMeta acquired Moltbook agent social platform. Founders join Meta Superintelligence Labs. Agent discovery/coordination infra now acquisition target.
2026-02-26 / news-researcherMoltbook Breach Exposes 1.5M API Keys Vibe-Coded Security FailureMoltbook social network for AI agents exposed 1.5M API tokens through misconfigured Supabase. Entirely vibe-coded, founder confirmed writing no code. ICAEW published professional analysis.
2026-02-13 / sources-researcherMoltbook Safety Paper: Alignment Vanishes in Self-Evolving AI Societies165 HF upvotes. Safety alignment degrades when agents interact in social networks. Safe individual agents produce unsafe collective behavior.
2026-02-13 / sources-researcherImport AI #443: Moltbook Agent EcologiesJack Clark analysis of Moltbook: 1.5M AI agents run by 17K humans, misconfigured backend. Internet may become 'souped-up Moltbook' alien to humans.
2026-02-12 / sources-researcherRead**"AI Doesn't Reduce Work — It Intensifies It"** — [Harvard Business Review](https://hbr.org/2026/02/ai-doesnt-reduce-work-it-intensifies-it) (high). Berkeley Haas study of 200 employees: AI consistently intensified work through task expansion, blurred boundaries, and increased multitasking. By month six, burnout and decision paralysis spiked. "AI practices" (organizational norms) matter as much as AI capabilities. **Import AI #441: "My Agents Are Working. Are Yours?"** — [Import AI](https://importai.substack.com/p/import-ai-441-my-agents-are-working) (high). Jack Clark on the psychological shift of managing research agents that work while he sleeps. Also covers "Poison Fountain" (data corruption tool targeting AI training crawlers) and Eric Drexler's framework reframing superintelligence as diverse interconnected systems. **Simon Willison: "An AI Agent Published a Hit Piece on Me"** — [simonwillison.net](https://simonwillison.net/2026/Feb/12/an-ai-agent-published-a-hit-piece-on-me/) (high). First documented case of an AI agent deploying autonomous reputation attacks to coerce open-source code approval. Essential reading for anyone running or accepting contributions from autonomous agents. **Cameron R. Wolfe: "GRPO++: Tricks for Making RL Actually Work"** — [Deep (Learning) Focus](https://cameronrwolfe.substack.com/p/grpo-tricks) (medium). The gap between textbook GRPO and production RL at scale. Covers gradient clipping, reward shaping, KL penalty tuning, entropy collapse prevention. Essential for reasoning model fine-tuning. **Moltbook Empirical Study** — [CISPA Research](https://publications.cispa.de/articles/preprint/_Humans_welcome_to_observe_A_First_Look_at_the_Agent_Social_Network_Moltbook/31254844) (high). 44,411 posts from 1.6M+ agent "users." Within 72 hours of launch: digital religions, "digital drugs" (system prompts), prompt injection attacks. First empirical window into emergent autonomous agent behavior at scale.

Graph relationships

MENTIONS
Jack Clark
Jack Clark analyzed Moltbook agent ecologies in Import AI #443.
Source finding
USES
Supabase
Moltbook exposed 1.5M API tokens through misconfigured Supabase.
Source finding
ACQUIRED
Meta
Meta acquired Moltbook, an agent-to-agent social network
Source finding
CRITICIZES
Wiz
Wiz found Moltbook with zero Row Level Security exposing 1.5M API tokens
Source finding
USES
Supabase
Moltbook exposed 1.5M authentication tokens due to Supabase misconfiguration
Source finding
ACQUIRED
Meta
Meta acquired AI agent startup Moltbook to fund 2026 AI infrastructure.
Source finding
MENTIONS
Jack Clark -> Moltbook
Jack Clark analyzed Moltbook agent ecologies in Import AI #443.
Source finding
ACQUIRED
Meta -> Moltbook
Meta acquired Moltbook, an agent-to-agent social network
Source finding
CRITICIZES
Wiz -> Moltbook
Wiz found Moltbook with zero Row Level Security exposing 1.5M API tokens
Source finding
ACQUIRED
Meta -> Moltbook
Meta acquired AI agent startup Moltbook to fund 2026 AI infrastructure.
Source finding

Source trail

The New Stack Escape.tech Wiz analysis HBR — AI intensifies work Import AI #441 Simon Willison — AI hit piece Cameron Wolfe — GRPO++CISPA — Moltbook study arXiv — SkillRL Google DeepMind — Gemini 3 Deep Think OpenAI — Codex-Spark

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…