Weekly briefing

SecurityPGHoard: Incomplete Path Traversal fix in restore.py allows Arbitrary File Write

SecurityPGHoard: Password written to debug log

SecurityPart-DB: Attachment Size Limit Bypass via API Base64 Upload

SecurityPart-DB: Authenticated Disk Exhaustion via Unbounded Server-Side Attachment Downloads

SecurityPart-DB: Hardcoded APP_SECRET enables forged remember-me cookie

SecurityPart-DB: Password Reset Link Poisoning via Host Header Injection leads to pre-authentication Account Takeover

SecurityPart-DB: SVG Sanitization Bypass via Extension Spoofing Leads to Stored XSS in Attachment Viewer

SecurityPart-DB: Spreadsheet Formula Injection in XLSX Part Exports

SecurityPart-DB: Stored HTML injection in BOM validation results via unescaped Part name

SecurityPart-DB: Stored XSS / HTML injection in activity-log Extra column via unsanitized CollectionElementDeleted element name (sibling-sanitizer gap)

SecurityPopup Builder < 4.4.6 - Unauthenticated Popup Trashing via post_type Parameter

SecurityPraisonAI A2U incomplete authentication fix leaves current serve command unauthenticated by default

SecurityPraisonAI DiscordApproval accepts unrelated channel messages as dangerous-tool approvals

SecurityPraisonAI Dynamic Context history and terminal tools read files outside configured storage via path traversal

SecurityPraisonAI GitHub template cache path traversal allows outside-cache file write and directory deletion

SecurityPraisonAI LinearBot processes unsigned webhooks when LINEAR_WEBHOOK_SECRET is missing

SecurityPraisonAI SandlockSandbox falls back to unrestricted subprocess execution when Landlock is unavailable

SecurityPraisonAI Slack app_mention bypasses configured user/channel authorization

SecurityPraisonAI dynamic-context artifact tools read arbitrary host files outside artifact storage

SecurityPraisonAI recipe serve Typer command bypasses the non-localhost authentication guard

SecurityPraisonAI recipe workflow policy can be bypassed by declaring and YAML-approving dangerous tools outside TEMPLATE.yaml

SecurityPraisonAI recipe.run_stream skips dangerous-tool policy enforcement

SecurityPraisonAI: Compute-bridged file tools allow shell command injection

SecurityPraisonAI: HTTPApproval dashboard renders tool arguments as raw HTML, allowing approval-page XSS to approve dangerous tools