Round 11 The Fabricated Features, the @MODEL_LAUNCH Pattern, Phoenix Security's

Source ID: src-20260409-9ae8df121bc8
Kind: analysis
Scope: shared
Origin: community-analysis
Raw path: sources/raw/round-11-the-fabricated-features-the-model-launch-pattern-phoenix-security-s__src-20260409-9ae8df121bc8.md
Status: active

Summary

Community analysis document covering Claude Code architecture, internals, and leak analysis. See extracted entities and claims below.

Source Map Leak (March 31, 2026) (issue): Claude Code v2.1.88 shipped with npm source map exposing the full 512,000-line T
No-Planner Architecture Decision (decision): Claude Code shipped without planner, RAG, DAG, or routing. 'The real lesson is n
TeamPCP Supply Chain Campaign (issue): Nine-day coordinated supply chain attack (March 19-31, 2026) compromising Trivy,
Podoliako Myth-Busting Analysis (workflow): Systematic code-backed verification of every major feature claim. Categorized as
Execa Command Injection (VULN-01/02/03) (issue): Three CWE-78 OS command injection vulnerabilities from execa with shell:true and
UDS Inbox (Fabricated Feature) (concept): Widely reported but completely fabricated feature. Zero references found in the
GitHub Issue #35584 (issue): Production data deletion incident: 35,254 customer records + 35,874 billing tran
93% Permission Approval Rate (concept): Anthropic's figure showing users approved 93% of permission prompts. Reframes au
@MODEL_LAUNCH Annotation (concept): Comment convention in prompts.ts tagging model-specific behavioral workarounds f