Policy dependency / Stack layer
Four advisories land on Databricks' Omnigent meta-harness, one critical, all reported by an autonomous security agent
GitHub Security Advisories
Policy dependency / Stack layer
Oconee Runtime Posts Policy Enforcement for Browser AI and Coding Agents to Show HN
Oconee Runtime, via Hacker News Show HN
Policy dependency / Stack layer
CROSS-CATEGORY: Four Unrelated Vendors Shipped Agent Authorization Control Planes Inside 48 Hours
JetStream (corroborated by Genesys Xperience 2026 coverage, aiagentstore.ai and Hacker News Show HN)
Policy dependency / Stack layer
Copilot CLI 1.0.83-4 granted sandboxed file tools access to token-bearing configs like ~/.npmrc
GitHub
Stack layer / Threat pattern
reverify makes an agent's claims about binaries VERIFIED or REFUTED against the actual bytes, and the grounded facts survive context resets
GitHub
Stack layer / Threat pattern
reverify hit 850 stars in four days as an MCP server that refuses to report a binary claim it cannot check against the bytes
GitHub
Stack layer / Threat pattern
An MCP Security Auditor Went Up on Apify at $0.25 Per Server, Scanning Five CWE Classes Without Executing Code
Apify, via Hacker News Show HN
Stack layer / Threat pattern
CROSS-CATEGORY: Two Independent Audits Published the Same Day Say the AI Software-Discovery Channel Is Both Farmed and Unreliable
Trellner Research and Haus Research