changelog
01Cursor rule Cursor
Update CHANGELOG.md for user-visible changes (Keep a Changelog).
32 tagged ai safety, measured the same way as everything else here.
Browse within: ai-governance 12Guardrails 7Autonomous Agents 6ai-orchestration 6enterprise-ai 5gemini 5
Cursor rule Cursor
Update CHANGELOG.md for user-visible changes (Keep a Changelog).
Cursor rule Cursor
Regenerate CRD manifests when Go types that define CRD schemas change.
Cursor rule Cursor
Keep OpenAPI in sync when the v1 API or CLI-exposed HTTP surface changes.
elliot35/deterministic-agent-control-protocol
Cursor rule
Enforce governance on all file and system operations.
Cursor rule Cursor
P8 Enforcement Rules — Injected into every Agent conversation.
Cursor rule Cursor
Before telling the user a task is done, you MUST.
Cursor rule Cursor
把 agent-guardrails 作为 repo-local guardrail 层。.
Cursor rule
Before telling the user a task is done, you MUST.
Cursor rule
After any mistake or unexpected behavior, prompt to capture structured feedback with context, history-aware lesson distillation, and tags.
Cursor rule
Before risky tool calls (git push, rm -rf, npm publish, deploy), check prevention rules via the thumbgate MCP server. If a rule matches, STOP and explain why.
Cursor rule
At session start, recall past context. At session end, hand off state for next session.
Cursor rule Cursor
This project uses Phantom Secrets to protect API keys from AI leaks.
Cursor rule Cursor
A rule that requires generated code to pass through a review service before it is shown to the user.
Cursor rule Cursor
Cursor local adapter — CLI integration and skills sync.
Cursor rule Cursor
TitanClip pnpm monorepo layout, scripts, and scoping.
Cursor rule Cursor
Shared packages — db, shared, adapters, plugin SDK.
Cursor rule Cursor
Rules for working with the agent core loop in selectools.
Cursor rule Cursor
Selectools development conventions, testing requirements, and release process.
Cursor rule Cursor
Testing conventions and patterns for selectools.
Cursor rule Cursor
Reviews Python code for unguarded AI agent tool calls. Use when reviewing agent code, scanning for security issues, or checking tool call safety.
Cursor rule Cursor
When a request is ambiguous, high-impact, or potentially destructive, follow the "Risk Triage -> Align -> Act" protocol defined in skill.md.
Cursor rule Cursor
AgentGuard local-first safety.
Cursor rule Cursor
When working with AI agents in this project, use Authensor for safety.
Cursor rule Cursor
README.md # 唯一放在根目录的文档(项目入口) .cursorrules # 本文件 docs/ ├── design/ # 设计文档(architecture.md / data-model.md) ├── api/ # API 文档(api-reference.md) ├── specs/ # 功能技术规格(SPEC-NNN-.md,含 SPEC-005 IPC 权威源) ├── guides/ # 使用指南(development.md / deployment.md) ├── changelog/ # 变更日志(CHANGELOG.md) ├── research/ # 调研材料.