InnerWarden

InnerWarden/inner-warden

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

12 good first / help-wanted issues · Rust · last activity Oct 23, 2026

2 stars 1 forks 2 watchers Rust Apache License 2.0
ai-agents ai-security claude-code devsecops guardrails llm-security mcp prompt-injection rust sandbox
View on GitHub Website
12 Open Issues Need Help Last updated: Oct 23, 2026

Open Issues Need Help

View All on GitHub
Dashboard: put the current page name in the tab title 2 days ago
enhancement good first issue dashboard
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Dashboard: announce Copied to screen readers 2 days ago
enhancement good first issue dashboard
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
CONTRIBUTING: list every gate CI runs, with its local command 2 days ago
documentation good first issue
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
README: a first-commands section, and fix the CLI crate's README 2 days ago
documentation good first issue
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Assert each rule's YAML test_cases 2 days ago
help wanted detection tests
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Add benchmark corpus cases for the thin attack categories 2 days ago
good first issue detection tests
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Let install accept the agent names agents connect accepts 2 days ago
enhancement good first issue rust
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Refuse unknown flags in serve, graph, status and llm 2 days ago
bug good first issue rust
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Suggest the nearest command on a typo 2 days ago
enhancement good first issue rust
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
innerwarden check with no command on a terminal waits silently 2 days ago
enhancement good first issue rust
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Make four Unix-assuming unit tests pass on Windows 2 days ago
good first issue tests windows
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox
Check out text files with LF on Windows so pinned digests hold 2 days ago
bug good first issue tests windows
InnerWarden/inner-warden
2

Open-source runtime guardrail for AI agents: screens shell commands and MCP/tool calls before they run (allow/review/deny). Cross-OS, Apache-2.0.

Rust
#ai-agents#ai-security#claude-code#devsecops#guardrails#llm-security#mcp#prompt-injection#rust#sandbox