feat(capabilities): add dynamic context injection via UserPromptSubmit hooks anthropics/claude-cookbooks #675 opened 2 months ago by daveCode-dot 3 1
feat(lib): add strip_thinking_blocks utility for conversation history recovery anthropics/anthropic-sdk-python #1605 opened 2 months ago by daveCode-dot 3 1
fix(skills): auto-prefix upload paths with skill name directory anthropics/anthropic-sdk-python #1604 opened 2 months ago by daveCode-dot 3 2
[BUG] Opus 4.8 refusal on benign code-audit prompt — false positive Usage Policy block anthropics/claude-code #65175 opened 2 months ago by daveCode-dot bug platform:macos area:model 2
Tool inputSchema generation emits $ref for reused Zod instances · breaks strict MCP clients (kimi) modelcontextprotocol/typescript-sdk #2100 opened 2 months ago by daveCode-dot 3
Cyber safeguard: sticky per-session false-positive blocks (1 events, 1 request IDs) anthropics/claude-code #74489 opened 2 weeks ago by daveCode-dot bug area:model area:security
Cyber safeguard: sticky per-session false-positive blocks (28 events, 23 request IDs) across Opus 4.8 + Sonnet 5 anthropics/claude-code #74430 opened 2 weeks ago by daveCode-dot bug area:model area:security 1
Quota consumption on Max 20x plan feels disproportionate to actual usage anthropics/claude-code #72579 opened 3 weeks ago by daveCode-dot question area:cost 1
False positive: API safeguards block legitimate security code reviews via custom subagents [skip-publish-guard] anthropics/claude-code #72256 opened 3 weeks ago by daveCode-dot bug platform:macos area:model 1