Repository Issue Activity (beta)

#114 Plugin manifest at .plugin/plugin.json is not read — skills are namespaced by commit SHA

#105 Hosted-agents Modal example uses shell injection and unsafe secret handling

#106 API loops and judge examples lack cost / rate-limit safeguards

#107 Research loop fails on Windows due to Unix-only fcntl import

#108 Stale researcher-OS baseline numbers in README.md and AGENTS.md

#104 Project-development pipeline is vulnerable to path traversal / arbitrary file deletion

#42 interleaved_thinking - Skill directory name is not a valid OpenCode skill name

#99 Docs idea: quick first run for Context Engineering Skills

#93 Add a runtime-continuity / state-handoff skill for crash-resume boundaries

#35 Proposal: Add context-benchmarking Skill and CI Example

#91 Zip file contains path with invalid characters

#2 Add meta-skill: skill-usage-fundamentals (Using Skills Wisely)

#84 NLPM Audit: 4 bugs and 3 security improvements found (score: 83/100)

#79 Possible complementary direction: long-term skill lifecycle management with SkillClaw

#67 Landing page for Agent-Skills-for-Context-Engineering repo

#34 I see every plugin installed in marketplace is same

#65 Payment Execution Skill: AgentPay MCP for Agentic Commerce

#54 Add translate-image skill — AI image translation, OCR, and text removal

#45 Fix typos in documentation

#43 Verific

#18 Marketplace skill discovery broken: individual skills not discoverable

#1 Warehouse cooperation

#3 Convert to claude plugin