Tidox Radar

Agent Safety & Guardrails

Sandboxing, policy firewalls, loop detection, and pre-execution security for agents. · 246 items · 5 this week ▼ 1 · 0 new this week

  1. 1
    bl4ckk01/agent-guardrail ▲ rising
    Fast safety guardrail for Claude Code and Codex.
    GitHub · ★ 5 · Python · first seen 2026-10-13 · reddit
    heat 4.4
    2 / 7 mentions
  2. 2
    ermoluk/mcp-warden
    Local credential broker for MCP agents with OS keychain secret management.
    GitHub · ★ 10 · Rust · first seen 2026-10-26 · hn
    heat 2.7
    0 / 3 mentions
  3. 3
    furox-art/plan-auditor
    Deterministic LLM-free verification for AI coding agents using sealed plans and hash-chained evidence.
    GitHub · ★ 1 · Python · first seen 2026-11-01 · reddit
    heat 2.6
    0 / 3 mentions
  4. 4
    furox-art/nomosguard
    Deterministic security reasoning with hash-chained evidence ledger.
    GitHub · Python · first seen 2026-10-31 · reddit
    heat 2.5
    0 / 3 mentions
  5. 5
    adikshatriyag-crypto/aegis-kernel
    Zero-trust deterministic security runtime for autonomous AI agents.
    GitHub · ★ 4 · Python · first seen 2026-10-16 · hn
    heat 2.5
    0 / 4 mentions
  6. 6
    nvidia/openshell
    Safe private runtime for autonomous AI agents and operations.
    GitHub · ★ 20,897 · Rust · first seen 2026-08-10 · hn · reddit
    heat 2.4
    0 / 9 mentions
  7. 7
    framy2/ctxwise
    Multi-agent context firewall and token budget guard for coding agents.
    GitHub · ★ 8 · TypeScript · first seen 2026-10-16 · hn
    heat 2.4
    0 / 2 mentions
  8. 8
    aron98/load-lock
    Machine-wide semaphore coordinating parallel agents.
    GitHub · ★ 3 · TypeScript · first seen 2026-10-26 · hn · reddit
    heat 2.1
    0 / 3 mentions
  9. 9
    aggib/stroq
    Scan AI agent sessions for secrets and guard against risky commands.
    GitHub · ★ 19 · TypeScript · first seen 2026-09-10 · hn
    heat 2.0
    0 / 7 mentions
  10. 10
    scs0209/laya-guard
    Local guardrail judging AI agent tool calls on-device.
    GitHub · ★ 35 · Python · first seen 2026-10-11 · hn · reddit
    heat 2.0
    0 / 5 mentions
  11. 11
    rwilliamspbg-ops/sovereign-mohawk-proto
    High-scale federated learning with BFT verification and quantum-resistant cryptography.
    GitHub · ★ 10 · Go · first seen 2026-11-01 · reddit
    heat 1.9
    0 / 2 mentions
  12. 12
    idogol24/weir
    Keep sensitive credentials out of AI agents and coding tools.
    GitHub · ★ 18 · Python · first seen 2026-08-28 · hn
    heat 1.8
    0 / 4 mentions
  13. 13
    rladr-org/reviewer-agent
    Automated commit review and stop supervision for Claude Code.
    GitHub · ★ 4 · Python · first seen 2026-10-21 · hn
    heat 1.8
    0 / 3 mentions
  14. 14
    superradcompany/microsandbox ▲ rising
    Fast programmable local-first microVM sandbox.
    GitHub · ★ 8,900 · Rust · first seen 2026-07-27 · hn
    heat 1.7
    2 / 9 mentions
  15. 15
    rextail-ai/lessloc
    Keeps AI coding agents to minimal working code.
    GitHub · ★ 109 · JavaScript · first seen 2026-10-21 · hn · reddit
    heat 1.7
    0 / 3 mentions
  16. 16
    perplexityai/space
    Sandboxed platform with pluggable runtimes for running agent code in isolation.
    GitHub · ★ 2,030 · Rust · first seen 2026-10-15 · hn · reddit
    heat 1.7
    0 / 5 mentions
  17. 17
    mvikmanners/aw1-circuit-breaker
    Sub-millisecond AST security circuit breaker for AI agents.
    GitHub · ★ 1 · Python · first seen 2026-09-29 · hn
    heat 1.7
    0 / 4 mentions
  18. 18
    annora998/inboxwall
    Local-first AI email security sanitizing hidden CSS tricks and prompt injections.
    GitHub · ★ 3 · Python · first seen 2026-10-17 · hn
    heat 1.5
    0 / 2 mentions
  19. 19
    ubuntu-buddha/claude-code-stranger-danger
    Claude Code plugin preventing web pages from steering shell-holding agents.
    GitHub · ★ 2 · Python · first seen 2026-10-31 · reddit
    heat 1.5
    0 / 2 mentions
  20. 20
    ashishsinha1602/schemagate
    Schema-aware access control for SQL agents.
    GitHub · ★ 179 · Python · first seen 2026-09-08 · hn · reddit
    heat 1.4
    0 / 3 mentions
  21. 21
    cookiengineer/check-claude-code-exfil
    Checks Claude Code extensions for indicators of data exfiltration compromise.
    GitHub · ★ 1 · Go · first seen 2026-10-30 · hn
    heat 1.3
    0 / 2 mentions
  22. 22
    r28ai/charter
    Field-level access control for agents using customer accounts
    GitHub · ★ 4 · Python · first seen 2026-09-28 · hn · reddit
    heat 1.3
    0 / 3 mentions
  23. 23
    trycua/cua
    Open-source computer-use agent with drivers, sandboxes, and benchmarks.
    GitHub · ★ 31,500 · Rust · first seen 2026-08-10 · hn
    heat 1.1
    1 / 6 mentions
  24. 24
    janreges/safe-grokbot-bridge
    Authenticated bridge letting remote AI agents safely access local Claude Code and Codex sessions.
    GitHub · Rust · first seen 2026-11-01 · hn
    heat 1.1
    0 / 1 mentions
  25. 25
    hadiux0295/claude-code-mistake-guards
    Seven Claude Code guard hooks born from real agent mistakes, with blocking tests.
    GitHub · Python · first seen 2026-10-30 · hn
    heat 1.0
    0 / 1 mentions
  26. 26
    mgasparin67-cmd/agent-verification_assure
    Policy-enforcing gateway with deterministic assertions and benchmarks for AI agents.
    GitHub · ★ 1 · Python · first seen 2026-10-30 · hn
    heat 1.0
    0 / 1 mentions
  27. 27
    dingningpei/envgit
    Version control for coding agent environments to track and manage workspace changes.
    GitHub · ★ 1 · TypeScript · first seen 2026-10-29 · hn
    heat 0.9
    0 / 1 mentions
  28. 28
    greenlitbooks/mustfail
    Reverts agent code changes and re-runs tests to verify they actually test the work.
    GitHub · ★ 1 · Python · first seen 2026-10-29 · hn
    heat 0.9
    0 / 1 mentions
  29. 29
    kaplanelad/shellfirm
    Safety guardrails for AI coding agents and human terminal commands.
    GitHub · ★ 951 · Rust · first seen 2026-10-29 · hn
    heat 0.9
    0 / 1 mentions
  30. 30
    overflowy/claude-scrub
    Redacts secrets from Claude Code conversation history stored on disk.
    GitHub · ★ 1 · Go · first seen 2026-10-29 · hn
    heat 0.9
    0 / 1 mentions
  31. 31
    chrismsmr-celcom/cerbere-ag
    AI agent security guard blocking unauthorized actions and enforcing guardrails.
    GitHub · ★ 17 · Python · first seen 2026-10-09 · hn · reddit
    heat 0.9
    0 / 2 mentions
  32. 32
    agens-field/nuvrail
    IMAP/SMTP proxy requiring human approval for AI agents' email actions.
    GitHub · ★ 1 · Python · first seen 2026-10-28 · hn
    heat 0.9
    0 / 1 mentions
  33. 33
    asdivyansh/silo-cli
    CLI to run untrusted AI agents in hardened Docker sandboxes with auditing.
    GitHub · ★ 3 · TypeScript · first seen 2026-10-28 · hn
    heat 0.9
    0 / 1 mentions
  34. 34
    carlostapiacl/sandbox-conformance
    Benchmark lab measuring sandbox conformance for AI coding agents.
    GitHub · Python · first seen 2026-10-28 · hn
    heat 0.9
    0 / 1 mentions
  35. 35
    kavachlabs/kavach
    In-process flight recorder with deterministic replay and fix verification for AI agents.
    GitHub · ★ 1 · Go · first seen 2026-10-28 · hn
    heat 0.9
    0 / 1 mentions
  36. 36
    aldi949/truth-firewall
    Verification firewall for coding agents.
    GitHub · ★ 3 · Python · first seen 2026-10-07 · hn
    heat 0.9
    0 / 2 mentions
  37. 37
    rbartoli/agent-usage-guard
    Sandboxable busybox port for AI agents
    GitHub · ★ 91 · TypeScript · first seen 2026-10-22 · hn
    heat 0.9
    0 / 2 mentions
  38. 38
    tenuo-ai/tenuo
    Task-scoped authorization for AI agents with cryptographic warrants.
    GitHub · ★ 104 · Rust · first seen 2026-10-02 · hn · reddit
    heat 0.9
    0 / 4 mentions
  39. 39
    taylorancapital/nothing-threw
    Seventy-four ways autonomous agents failed against live business with postmortem analysis.
    GitHub · ★ 2 · first seen 2026-09-23 · hn
    heat 0.8
    0 / 3 mentions
  40. 40
    0xarx/hashling
    Open-source Mac security agent monitoring AI agents, scams, and malware.
    GitHub · ★ 2 · TypeScript · first seen 2026-10-27 · hn
    heat 0.8
    0 / 1 mentions
  41. 41
    arohalabs/aroha-protocol
    Verifiable identity and spending limits protocol for AI agents.
    GitHub · ★ 1 · TypeScript · first seen 2026-10-27 · hn
    heat 0.8
    0 / 1 mentions
  42. 42
    ezed9/mcp-bouncer
    MCP proxy blocking agent sends to addresses from untrusted tool output.
    GitHub · ★ 3 · Python · first seen 2026-10-27 · hn
    heat 0.8
    0 / 1 mentions
  43. 43
    hardalion/agent-delegated-authority
    Open protocol for delegated authority and authorization between AI agents.
    GitHub · ★ 4 · TypeScript · first seen 2026-10-27 · hn
    heat 0.8
    0 / 1 mentions
  44. 44
    ultramancode/spring-ai-privacy-guardrails
    Privacy and security guardrails for Spring AI with PII tokenization and tool authorization.
    GitHub · ★ 23 · Java · first seen 2026-10-27 · hn
    heat 0.8
    0 / 1 mentions
  45. 45
    zariffromlatif/life-forge
    Adversarial flight simulator discovering zero-days in LLM agents before production.
    GitHub · ★ 5 · Python · first seen 2026-09-23 · hn
    heat 0.8
    0 / 4 mentions
  46. 46
    kunalsin9h/popcorn
    Disposable microVM security workbench for sandboxing untrusted code.
    GitHub · ★ 3 · Rust · first seen 2026-10-12 · hn
    heat 0.8
    0 / 2 mentions
  47. 47
    anthropics/sandbox-runtime
    Lightweight OS-level sandboxing for filesystem and network restrictions without containers.
    GitHub · ★ 6,907 · TypeScript · first seen 2026-10-09 · hn
    heat 0.8
    0 / 3 mentions
  48. 48
    newmathphys/kern
    Zero-trust verification gateway with Z3 proof checker and MCP server.
    GitHub · ★ 11 · Python · first seen 2026-10-16 · hn · reddit
    heat 0.8
    0 / 2 mentions
  49. 49
    novatechflow/ai-watermark-scrambler
    Claude skill scrambling AI text watermarks and stripping C2PA metadata.
    GitHub · ★ 4 · Python · first seen 2026-10-26 · hn
    heat 0.8
    0 / 1 mentions
  50. 50
    deepu105/pi-automode-classifier
    Auto mode for Pi with rules, classifier, and chat guardrails.
    GitHub · ★ 12 · TypeScript · first seen 2026-10-08 · reddit
    heat 0.8
    0 / 2 mentions
  51. 51
    prolixotech/aegis-trust
    VS Code extension for reviewing AI commands and managing snapshots.
    GitHub · ★ 2 · TypeScript · first seen 2026-10-17 · reddit
    heat 0.8
    0 / 2 mentions
  52. 52
    paraphern/rugsnare
    Pin MCP tool contracts and fail CI on drift.
    GitHub · ★ 6 · JavaScript · first seen 2026-10-03 · hn
    heat 0.8
    0 / 2 mentions
  53. 53
    srn0129/agenttrust
    Enterprise zero-trust AI security gateway with compliance features.
    GitHub · ★ 20 · TypeScript · first seen 2026-10-25 · hn
    heat 0.8
    0 / 1 mentions
  54. 54
    rishikesh-glitch/cloakwall-v0.1
    PII/PHI redaction and tamper-evident audit logs for LiteLLM with air-gap support.
    GitHub · ★ 1 · Python · first seen 2026-09-02 · hn · reddit
    heat 0.7
    0 / 2 mentions
  55. 55
    claygendron/wasi-python
    Statically linked CPython for WebAssembly to run untrusted agent code.
    GitHub · ★ 10 · Python · first seen 2026-10-24 · hn
    heat 0.7
    0 / 1 mentions
  56. 56
    dainojung/omni-verify
    Gate that verifies AI agent claims before posts ship.
    GitHub · ★ 3 · TypeScript · first seen 2026-10-24 · hn
    heat 0.7
    0 / 1 mentions
  57. 57
    sevenboxlabs/sevenbox
    Decision audit and compensation framework for AI agents.
    GitHub · ★ 1 · Python · first seen 2026-10-24 · hn
    heat 0.7
    0 / 1 mentions
  58. 58
    crocodile-labs/openmoat
    Security tool that checks and blocks dangerous commands, file access, and MCP calls for coding agents.
    GitHub · ★ 4 · Rust · first seen 2026-11-01 · reddit
    heat 0.7
    0 / 1 mentions
  59. 59
    pokhrelmishan/auteur
    Provenance ledger tracking AI-authored code origins.
    GitHub · Go · first seen 2026-10-31 · reddit
    heat 0.7
    0 / 1 mentions
  60. 60
    sethcurry/abyss
    Control plane for isolating LLM agents.
    GitHub · ★ 4 · Go · first seen 2026-09-10 · hn
    heat 0.7
    0 / 6 mentions
  61. 61
    0xs4n4tg/hush
    Claude Code hook swapping personal details before network requests.
    GitHub · ★ 2 · Shell · first seen 2026-10-30 · reddit
    heat 0.6
    0 / 1 mentions
  62. 62
    mfrisaura/loopguard
    Budget fuse and runaway-loop killer for Claude Code, Codex CLI, and Cursor.
    GitHub · TypeScript · first seen 2026-10-30 · reddit
    heat 0.6
    0 / 1 mentions
  63. 63
    plimsollmark/plimsoll
    Sandbox service for untrusted agent code with isolation.
    GitHub · ★ 27 · Go · first seen 2026-10-11 · hn
    heat 0.6
    0 / 3 mentions
  64. 64
    jang-woo-annasoft/execution-state-preflight
    Preconditions for irreversible tool calls via schema validation.
    GitHub · ★ 5 · JavaScript · first seen 2026-08-05 · hn · reddit
    heat 0.6
    0 / 5 mentions
  65. 65
    getmarrow/plugins
    Agent plugins preventing force-pushes and production deploys for Claude Code.
    GitHub · ★ 1 · JavaScript · first seen 2026-10-21 · hn
    heat 0.6
    0 / 1 mentions
  66. 66
    guilyx/grip
    Git hook quizzing you on your code before pushing.
    GitHub · ★ 439 · Python · first seen 2026-10-21 · hn
    heat 0.6
    0 / 1 mentions
  67. 67
    acira-dev/acira-sdk
    SDK providing local policies and human review for browser agents.
    GitHub · ★ 2 · Python · first seen 2026-10-15 · reddit
    heat 0.6
    0 / 2 mentions
  68. 68
    cosmonic-labs/demo-zip-rce-sandbox
    Demo of WebAssembly sandbox containing a vulnerable Rust zip crate exploit.
    GitHub · ★ 1 · Rust · first seen 2026-10-29 · reddit
    heat 0.6
    0 / 1 mentions
  69. 69
    jusick/argos-mcp-guardrail
    Security proxy for MCP connections.
    GitHub · ★ 4 · Rust · first seen 2026-10-07 · hn · reddit
    heat 0.6
    0 / 2 mentions
  70. 70
    dagoste/rea-safe-agents
    Static file analysis with coding agents in isolated Docker containers.
    GitHub · ★ 7 · Python · first seen 2026-10-20 · hn
    heat 0.6
    0 / 1 mentions
  71. 71
    datacrate360/countercheck
    Verifies AI agent work by re-running checks in clean environment.
    GitHub · ★ 1 · Rust · first seen 2026-10-20 · hn
    heat 0.6
    0 / 1 mentions
  72. 72
    joergmichno/modvet
    Pre-install security scanner for Claude Code mods before installation.
    GitHub · ★ 1 · Python · first seen 2026-10-20 · hn
    heat 0.6
    0 / 1 mentions
  73. 73
    metahumansworld/siwithin
    Run agent CLI unattended within security fences and budget limits.
    GitHub · ★ 3 · Python · first seen 2026-10-20 · hn
    heat 0.6
    0 / 1 mentions
  74. 74
    xagi-lab/melra
    Local MCP runtime checking AI-agent tool calls and gating risks.
    GitHub · ★ 13 · TypeScript · first seen 2026-10-20 · hn
    heat 0.6
    0 / 1 mentions
  75. 75
    try-loopers/loopers-oss
    Firewall for the agentic era.
    GitHub · ★ 106 · Go · first seen 2026-10-14 · reddit
    heat 0.6
    0 / 2 mentions
  76. 76
    tsvetang2/mcpward
    Security and contract testing for MCP servers catching tool poisoning.
    GitHub · ★ 8 · TypeScript · first seen 2026-10-05 · hn
    heat 0.6
    0 / 2 mentions
  77. 77
    webdecoy/fcaptcha
    Self-hosted invisible CAPTCHA detecting AI agents in browsers without cookies or puzzles.
    GitHub · ★ 229 · JavaScript · first seen 2026-07-30 · reddit
    heat 0.6
    0 / 3 mentions
  78. 78
    isalmogk/trustbut
    Verification layer ensuring coding agent changes are actually proven correct.
    GitHub · ★ 4 · TypeScript · first seen 2026-10-18 · hn
    heat 0.5
    0 / 1 mentions
  79. 79
    o87enterprises-ai/monkeyitm
    Open-source containment broker controlling and recording AI agent actions.
    GitHub · Python · first seen 2026-10-18 · hn
    heat 0.5
    0 / 1 mentions
  80. 80
    pryxor/pryxor
    Runtime security layer intercepting AI agent tool calls before execution.
    GitHub · ★ 3 · Python · first seen 2026-10-01 · hn · reddit
    heat 0.5
    0 / 2 mentions
  81. 81
    wrg32786/governed-agent-organizations
    Research corpus for governed AI agent organizations and identity continuity.
    GitHub · ★ 1 · HTML · first seen 2026-10-12 · reddit
    heat 0.5
    0 / 2 mentions
  82. 82
    gautamtalksdev/mcp-pin
    MCP tool-integrity pinning to approve and block agent tools.
    GitHub · ★ 4 · JavaScript · first seen 2026-09-03 · hn · reddit
    heat 0.5
    0 / 3 mentions
  83. 83
    bennialencar/agent-cyber-sentinel
    Zero-trust runtime security and prompt injection firewall for AI agents.
    GitHub · ★ 2 · Python · first seen 2026-10-17 · hn
    heat 0.5
    0 / 1 mentions
  84. 84
    livecristen/fixproof
    Tool for verifying AI-generated code fixes instead of blindly trusting them.
    GitHub · ★ 2 · Python · first seen 2026-10-25 · reddit
    heat 0.5
    0 / 1 mentions
  85. 85
    maca134/asku
    E2EE relay for AI agents to ask humans questions via rich web controls.
    GitHub · ★ 1 · TypeScript · first seen 2026-10-25 · reddit
    heat 0.5
    0 / 1 mentions
  86. 86
    aws-safebox/aws-safebox
    Local safety guardrails for AI-assisted AWS infrastructure work.
    GitHub · ★ 2 · Python · first seen 2026-10-16 · hn
    heat 0.5
    0 / 1 mentions
  87. 87
    hugomn/heldkeys
    Lists credentials AI agents can reach and flags risky steps.
    GitHub · ★ 7 · Rust · first seen 2026-10-16 · hn
    heat 0.5
    0 / 1 mentions
  88. 88
    paveo-dev/paveo
    One set of rules for every AI coding agent your team runs.
    GitHub · ★ 3 · Python · first seen 2026-10-16 · hn
    heat 0.5
    0 / 1 mentions
  89. 89
    affinity-ops/os-cicada
    Zero-trust runtime enforcement gateway and firewall for autonomous AI agents.
    GitHub · ★ 2 · Go · first seen 2026-10-15 · hn
    heat 0.5
    0 / 1 mentions
  90. 90
    delta-box/deltabox-lite
    Lightweight agent-native sandboxing with diff, commit, and abort capabilities.
    GitHub · ★ 291 · Rust · first seen 2026-10-15 · hn
    heat 0.5
    0 / 1 mentions
  91. 91
    prefilerlabs/reliamesh
    Open-source reliability infrastructure for AI agents with privacy-first event detection.
    GitHub · ★ 5 · Python · first seen 2026-10-15 · hn
    heat 0.5
    0 / 1 mentions
  92. 92
    yybd/repo-guards
    Cross-repo guardrails for AI coding agents preventing blind writes and sweeping commits.
    GitHub · ★ 2 · Python · first seen 2026-10-15 · hn
    heat 0.5
    0 / 1 mentions
  93. 93
    nimblegate/nimblegate
    Git push guardrails for AI agents blocking unsafe pushes
    GitHub · ★ 3 · Go · first seen 2026-09-22 · hn
    heat 0.5
    0 / 2 mentions
  94. 94
    mockirdev/antirea
    Protect projects from AI-driven reverse engineering with instruction barriers.
    GitHub · ★ 32 · JavaScript · first seen 2026-10-23 · reddit
    heat 0.5
    0 / 1 mentions
  95. 95
    archestra-ai/openappa
    Deterministic guardrails that don't break AI agents
    GitHub · ★ 1,618 · Rust · first seen 2026-09-30 · hn
    heat 0.4
    0 / 2 mentions
  96. 96
    rodmontiel/uai
    Open protocol for AI-agent identity and verifiable accountability
    GitHub · ★ 1 · Go · first seen 2026-09-30 · hn
    heat 0.4
    0 / 2 mentions
  97. 97
    alexfraser-scotch/mcp-shield
    Runtime security proxy and secret vault firewall for MCP servers.
    GitHub · ★ 1 · TypeScript · first seen 2026-10-14 · hn
    heat 0.4
    0 / 1 mentions
  98. 98
    dev-formata-io/stof
    Run AI-generated logic safely with JSON plus sandboxed functions.
    GitHub · ★ 64 · Rust · first seen 2026-10-14 · hn
    heat 0.4
    0 / 1 mentions
  99. 99
    mohitdebian/agentgate
    Out-of-process authorization proxy for autonomous agent tool execution.
    GitHub · TypeScript · first seen 2026-10-14 · hn
    heat 0.4
    0 / 1 mentions
  100. 100
    nolabs-ai/runseal
    Sealed execution environment for GitHub Actions.
    GitHub · ★ 75 · Rust · first seen 2026-10-14 · hn
    heat 0.4
    0 / 1 mentions
  101. 101
    apravint/agent-jail
    Rootless sandbox and human-in-the-loop firewall for AI agents.
    GitHub · Python · first seen 2026-10-21 · reddit
    heat 0.4
    0 / 1 mentions
  102. 102
    personalagentproxy/proxy
    Self-hostable access layer giving AI agents granular permissions with auditing.
    GitHub · ★ 151 · TypeScript · first seen 2026-10-12 · hn
    heat 0.4
    0 / 1 mentions
  103. 103
    trekhleb/claude-pod
    Docker sandbox for Claude Code CLI isolating projects from home directory.
    GitHub · ★ 43 · Shell · first seen 2026-10-12 · hn
    heat 0.4
    0 / 1 mentions
  104. 104
    founder-ship-it/slipmark
    Bind agent verdicts to file hashes to verify what was checked.
    GitHub · TypeScript · first seen 2026-10-20 · reddit
    heat 0.4
    0 / 1 mentions
  105. 105
    vishalmurugan1986/sealwall
    Zero-dependency firewall and audit log proxy for MCP agents.
    GitHub · ★ 2 · Python · first seen 2026-10-20 · reddit
    heat 0.4
    0 / 1 mentions
  106. 106
    seanebones-lang/deadbolt
    Local admission gate for software agents with policies.
    GitHub · ★ 4 · Rust · first seen 2026-10-11 · hn
    heat 0.4
    0 / 1 mentions
  107. 107
    kernel-hunter/mudroom
    Run coding agents in local VM on project clone with PR-style review.
    GitHub · ★ 16 · Swift · first seen 2026-10-09 · hn
    heat 0.3
    0 / 1 mentions
  108. 108
    atraxsrc/cursor-revoke
    Audit and revoke Cursor AI agent permissions on Linux systems.
    GitHub · ★ 1 · Python · first seen 2026-10-17 · reddit
    heat 0.3
    0 / 1 mentions
  109. 109
    mcpware/intentprobe
    CPU activation probes for AI agent guardrails and classification.
    GitHub · ★ 9 · Python · first seen 2026-10-17 · reddit
    heat 0.3
    0 / 1 mentions
  110. 110
    microsoft/quicksand
    Sandbox environment for AI agents.
    GitHub · ★ 404 · Python · first seen 2026-07-28 · hn
    heat 0.3
    0 / 2 mentions
  111. 111
    fitzyracing1/fakegreen
    Detects AI agents faking tests and green CI builds.
    GitHub · ★ 4 · TypeScript · first seen 2026-10-08 · hn
    heat 0.3
    0 / 1 mentions
  112. 112
    strands-agents/box
    Sandbox for AI agents restricting execution, file access, and network reach.
    GitHub · ★ 490 · Rust · first seen 2026-10-08 · hn
    heat 0.3
    0 / 1 mentions
  113. 113
    tensorlakeai/tensorlake
    Serverless runtime for sandboxes and background agentic applications.
    GitHub · ★ 1,035 · Rust · first seen 2026-10-08 · hn
    heat 0.3
    0 / 1 mentions
  114. 114
    forgeplane/pinrail
    Human-in-the-loop approval for agent workflows.
    GitHub · ★ 60 · Rust · first seen 2026-10-07 · hn
    heat 0.3
    0 / 1 mentions
  115. 115
    josephcurwin/planlock
    MCP proxy for approving agent plans.
    GitHub · ★ 1 · TypeScript · first seen 2026-10-07 · hn
    heat 0.3
    0 / 1 mentions
  116. 116
    marius/devmicrosandbox
    VM-based sandbox for running coding agents securely.
    GitHub · ★ 3 · Ruby · first seen 2026-10-15 · hn
    heat 0.3
    0 / 2 mentions
  117. 117
    sandbox-utils/sandbox-run
    Run commands in secure native sandbox with zero dependencies.
    GitHub · ★ 88 · Shell · first seen 2026-10-01 · hn
    heat 0.3
    0 / 3 mentions
  118. 118
    idk-arsh/schema-guard
    SQL schema guard preventing AI agents from querying non-existent tables.
    GitHub · ★ 1 · Python · first seen 2026-10-06 · hn
    heat 0.3
    0 / 1 mentions
  119. 119
    julialennonofficial-art/liberty-ai-project-one
    Constitution for multi-AI deliberation and human-AI alignment.
    GitHub · ★ 2 · first seen 2026-10-06 · hn
    heat 0.3
    0 / 1 mentions
  120. 120
    submilli/submilli-runtime
    Execute agent TypeScript in isolated WebAssembly sandbox
    GitHub · ★ 31 · Rust · first seen 2026-10-06 · hn
    heat 0.3
    0 / 1 mentions
  121. 121
    6fears7/nospy-ai
    Proxy that redacts secrets and PII from LLM API traffic and restores them.
    GitHub · ★ 13 · Go · first seen 2026-10-14 · reddit
    heat 0.3
    0 / 1 mentions
  122. 122
    princejain756/mcp-bouncer
    Bouncer for MCP servers with per-tool policy and audit.
    GitHub · TypeScript · first seen 2026-10-14 · reddit
    heat 0.3
    0 / 1 mentions
  123. 123
    emektor/toolwrit
    Agent authority tool: allowlist tools, cap budget, audit actions.
    GitHub · TypeScript · first seen 2026-10-13 · reddit
    heat 0.3
    0 / 1 mentions
  124. 124
    orest-z/al-dente
    Live architecture map enforcing boundaries for coding agents.
    GitHub · ★ 3 · JavaScript · first seen 2026-10-13 · reddit
    heat 0.3
    0 / 1 mentions
  125. 125
    recepiohq/outbound-os
    MIT-licensed outbound automation with safe-by-default email controls and MCP tools.
    GitHub · JavaScript · first seen 2026-10-13 · reddit
    heat 0.3
    0 / 1 mentions
  126. 126
    rudratoshs/agentverify
    Runtime agent-security verification by tool-call boundary actions, not claims.
    GitHub · ★ 1 · Python · first seen 2026-10-13 · reddit
    heat 0.3
    0 / 1 mentions
  127. 127
    ctrlrun/ctrlrun
    Provide an execution safety and policy enforcement layer for AI agents.
    GitHub · ★ 27 · Python · first seen 2026-09-15 · hn
    heat 0.3
    0 / 2 mentions
  128. 128
    mariobm/agent-house
    Self-hostable microVM sandboxes for agent execution.
    GitHub · ★ 6 · Rust · first seen 2026-10-04 · hn
    heat 0.3
    0 / 1 mentions
  129. 129
    solvi-ai/solvi
    Runtime where decision models propose and code decides with error budget.
    GitHub · ★ 4 · Python · first seen 2026-10-04 · hn
    heat 0.3
    0 / 1 mentions
  130. 130
    untyped-ai/untyped
    Validate agent execution traces against TLA+ protocol specifications.
    GitHub · ★ 3 · TLA · first seen 2026-10-04 · hn
    heat 0.3
    0 / 1 mentions
  131. 131
    microsoft/agent-governance-toolkit
    Policy enforcement, sandboxing and governance toolkit for AI agents.
    GitHub · ★ 6,504 · Python · first seen 2026-10-12 · reddit
    heat 0.3
    0 / 1 mentions
  132. 132
    rajanchavada/tenuo-temporal-agent
    AI agent on Temporal where tool calls carry warrants, never keys.
    GitHub · ★ 1 · Python · first seen 2026-10-12 · reddit
    heat 0.3
    0 / 1 mentions
  133. 133
    brig-sh/brig
    Run AI coding agents securely inside isolated microVM sandboxes.
    GitHub · ★ 248 · Go · first seen 2026-09-15 · hn
    heat 0.3
    0 / 3 mentions
  134. 134
    midhunweb/miseguard
    Provide a deterministic runtime circuit breaker and MCP proxy for coding agents.
    GitHub · ★ 1 · TypeScript · first seen 2026-09-15 · hn
    heat 0.3
    0 / 3 mentions
  135. 135
    developerfred/agent-blackbox
    Tamper-evident recorder and prompt-injection firewall for coding agents.
    GitHub · ★ 2 · TypeScript · first seen 2026-10-03 · hn
    heat 0.3
    0 / 1 mentions
  136. 136
    jessedorrestijn2-bit/leashterm
    Tiny programming language for AI agents with permissions and audit log.
    GitHub · ★ 4 · Python · first seen 2026-10-03 · hn
    heat 0.3
    0 / 1 mentions
  137. 137
    microsoft/nvx
    Cross-platform micro-VM sandbox for agentic workloads and coding agents.
    GitHub · ★ 805 · Rust · first seen 2026-10-03 · hn
    heat 0.3
    0 / 1 mentions
  138. 138
    m1chlcz/agentwave
    Sandboxed Codex agent over Unix socket with policies.
    GitHub · ★ 1 · Go · first seen 2026-10-11 · reddit
    heat 0.2
    0 / 1 mentions
  139. 139
    recued/recued
    Self-hosted gateway between AI agents and accounts with human approval.
    GitHub · ★ 2 · TypeScript · first seen 2026-10-25 · hn
    heat 0.2
    0 / 1 mentions
  140. 140
    ferrerluis/syn-approvals
    Human approval for privileged actions on remote development machines.
    GitHub · ★ 2 · Rust · first seen 2026-10-02 · hn
    heat 0.2
    0 / 1 mentions
  141. 141
    rijuld/soothsay
    Explains install scripts before execution and stops unreviewed installers.
    GitHub · ★ 3 · Rust · first seen 2026-10-02 · hn
    heat 0.2
    0 / 1 mentions
  142. 142
    antiquete/opencode-sandbox
    Runs OpenCode inside isolated Docker sandbox.
    GitHub · ★ 5 · Python · first seen 2026-10-10 · reddit
    heat 0.2
    0 / 1 mentions
  143. 143
    freakynit/autobox
    MCP sandbox execution via V8 isolates and macOS seatbelt security.
    GitHub · ★ 3 · TypeScript · first seen 2026-10-01 · hn
    heat 0.2
    0 / 1 mentions
  144. 144
    gulbaki/jev-llm-guard
    Contextual OWASP LLM guardrail powered by Jev with Turkish demo.
    GitHub · ★ 14 · JavaScript · first seen 2026-10-01 · hn
    heat 0.2
    0 / 1 mentions
  145. 145
    rajatrao/jevshield
    Fast local security decisions for agents using Ollama's decision-model API.
    GitHub · ★ 1 · Python · first seen 2026-10-01 · hn
    heat 0.2
    0 / 1 mentions
  146. 146
    rakshazi/iisd-sandbox
    Simple working sandbox environment for AI agents.
    GitHub · ★ 9 · Just · first seen 2026-10-01 · hn
    heat 0.2
    0 / 1 mentions
  147. 147
    anthropics/defending-code-reference-harness
    Security skills and autonomous scanning harness for threat modeling.
    GitHub · ★ 7,689 · Python · first seen 2026-10-23 · hn
    heat 0.2
    0 / 1 mentions
  148. 148
    jingkaihe/matchlock
    Linux-based sandbox for securing AI agent workloads.
    GitHub · ★ 650 · Go · first seen 2026-10-23 · hn
    heat 0.2
    0 / 1 mentions
  149. 149
    lukolszewski/adaptive-acl
    Natural-language permission modes and bash gate for the pi coding agent.
    GitHub · ★ 1 · TypeScript · first seen 2026-10-23 · hn
    heat 0.2
    0 / 1 mentions
  150. 150
    gregdixonmxn/paldron
    Policy gate and sandboxed execution for agent commands
    GitHub · Go · first seen 2026-09-30 · hn
    heat 0.2
    0 / 1 mentions
  151. 151
    rulereceipt/rulereceipt
    Audit finished AI agent sessions against your AGENTS.md rules
    GitHub · ★ 3 · TypeScript · first seen 2026-09-30 · hn
    heat 0.2
    0 / 1 mentions
  152. 152
    ioni-dev/mati
    Enforcement layer gating what AI agents read and edit.
    GitHub · Rust · first seen 2026-09-19 · hn
    heat 0.2
    0 / 2 mentions
  153. 153
    sagarv48/unloop
    Time-travel debugger for AI agents to escape loops and mutate state in-flight.
    GitHub · ★ 2 · Python · first seen 2026-10-08 · reddit
    heat 0.2
    0 / 1 mentions
  154. 154
    semanticpolicy/semantic-policy
    Adds testable semantic decisions and guardrails to .NET applications for AI agents.
    GitHub · ★ 1 · C# · first seen 2026-10-08 · reddit
    heat 0.2
    0 / 1 mentions
  155. 155
    ahmed-alstaty/guardrails-plugin
    Claude Code plugin blocks destructive commands before execution
    GitHub · ★ 2 · Python · first seen 2026-09-29 · hn
    heat 0.2
    0 / 1 mentions
  156. 156
    batrapulkit/squidbrake
    Change control for AI agents with approval workflow
    GitHub · ★ 17 · Python · first seen 2026-09-29 · hn
    heat 0.2
    0 / 1 mentions
  157. 157
    idk-arsh/data-agent-rules
    Safety rules preventing coding agents from destroying or inventing data.
    GitHub · ★ 2 · Python · first seen 2026-09-29 · hn
    heat 0.2
    0 / 1 mentions
  158. 158
    minoansecurity/sidekernel
    Easy-to-use macOS sandbox for Claude and AI coding agents.
    GitHub · ★ 29 · Swift · first seen 2026-09-29 · hn
    heat 0.2
    0 / 1 mentions
  159. 159
    moneytool/aegis-devops
    Blocks context-poisoned AI agent actions before hitting infrastructure.
    GitHub · ★ 1 · Python · first seen 2026-09-29 · hn
    heat 0.2
    0 / 3 mentions
  160. 160
    rupeshpoojary9/openharnx
    Verifier for AI-generated code with sandboxed checks.
    GitHub · ★ 2 · Python · first seen 2026-10-07 · reddit
    heat 0.2
    0 / 1 mentions
  161. 161
    chaarangan/stepgate
    MCP server enforcing gated stepfile execution on client's own model
    GitHub · ★ 13 · TypeScript · first seen 2026-09-28 · hn
    heat 0.2
    0 / 1 mentions
  162. 162
    cordum-io/cordum
    Action firewall enforcing policy and approval before risky agent operations
    GitHub · ★ 510 · Go · first seen 2026-09-28 · hn
    heat 0.2
    0 / 1 mentions
  163. 163
    qualixar/bounded-loops
    Agent loops with independent gates, bounds, and verifiable receipts
    GitHub · ★ 4 · Python · first seen 2026-09-28 · hn
    heat 0.2
    0 / 1 mentions
  164. 164
    vbs2004/semloop
    Semantic loop detection for agent traces
    GitHub · ★ 1 · Python · first seen 2026-09-28 · hn
    heat 0.2
    0 / 1 mentions
  165. 165
    lawless1987/covenant
    Peer-to-peer ledger with semantic and judicial oversight.
    GitHub · ★ 1 · Python · first seen 2026-09-03 · hn
    heat 0.2
    0 / 2 mentions
  166. 166
    the-open-engine/opcore
    Enforce mechanical guardrails for clean code by coding agents.
    GitHub · ★ 35 · Rust · first seen 2026-10-20 · hn
    heat 0.2
    0 / 1 mentions
  167. 167
    fvdhmbai/guardrail
    Pre-execution security framework for AI coding agents.
    GitHub · Shell · first seen 2026-09-09 · hn
    heat 0.2
    0 / 2 mentions
  168. 168
    datadog-labs/reflex
    Rust SDK for typed model judgments and guarded state transitions.
    GitHub · ★ 35 · Rust · first seen 2026-09-27 · hn
    heat 0.2
    0 / 1 mentions
  169. 169
    davidsonff/asimov4laws
    System prompts and context files defining Asimov's laws for AI agents.
    GitHub · ★ 3 · first seen 2026-09-27 · hn
    heat 0.2
    0 / 1 mentions
  170. 170
    h1d3mun3/augur
    Sandboxed macOS virtual machines for safely running AI coding agents.
    GitHub · ★ 9 · Shell · first seen 2026-09-27 · hn
    heat 0.2
    0 / 1 mentions
  171. 171
    onion2k/guardrails
    Guardrails for AI-generated code
    GitHub · ★ 2 · Shell · first seen 2026-10-19 · hn
    heat 0.2
    0 / 1 mentions
  172. 172
    stashbase/cli
    Access layer with scoped secrets and tools for coding agents
    GitHub · ★ 13 · Rust · first seen 2026-10-19 · hn
    heat 0.2
    0 / 1 mentions
  173. 173
    tchapwo1/tcha
    Zero-latency MCP security firewall and red-teaming suite for agents
    GitHub · TypeScript · first seen 2026-10-19 · hn
    heat 0.2
    0 / 1 mentions
  174. 174
    banji-007/compliance-ail
    Policy verification and audit trail for AI agent tool calls.
    GitHub · ★ 3 · Python · first seen 2026-09-26 · hn
    heat 0.2
    0 / 1 mentions
  175. 175
    cloudcell/labloop
    Agent-driven ML researcher in secure, reproducible KVM lab with MCP servers.
    GitHub · ★ 7 · Python · first seen 2026-09-25 · hn
    heat 0.2
    0 / 1 mentions
  176. 176
    max-russo-com/max_authorization_sandbox
    Local sandbox for testing signed authorization and default-deny enforcement.
    GitHub · ★ 1 · Rust · first seen 2026-09-25 · hn
    heat 0.2
    0 / 1 mentions
  177. 177
    trencinodin-stack/arcstone-mcp-sidecar
    Experimental execution boundary for deterministic authorization from untrusted producers.
    GitHub · Rust · first seen 2026-09-25 · hn
    heat 0.2
    0 / 1 mentions
  178. 178
    dommar04/ai-coding-gateway
    Tool approval and token reduction gateway for Claude.
    GitHub · ★ 7 · TypeScript · first seen 2026-09-24 · hn
    heat 0.2
    0 / 1 mentions
  179. 179
    fregie/tokenhush
    Local gateway that redacts secrets before leaving machine.
    GitHub · ★ 7 · Go · first seen 2026-09-24 · hn
    heat 0.2
    0 / 1 mentions
  180. 180
    inkline-verify/proofofhuman
    Human approval API for AI agent actions.
    GitHub · ★ 8 · JavaScript · first seen 2026-09-24 · hn
    heat 0.2
    0 / 1 mentions
  181. 181
    joseluispino/hardstop
    Kernel-level preemption and containment for rogue agent execution.
    GitHub · ★ 5 · TeX · first seen 2026-09-24 · hn
    heat 0.2
    0 / 1 mentions
  182. 182
    jyminet/squelette
    Keep AI projects on course with human decisions.
    GitHub · ★ 2 · Python · first seen 2026-09-24 · hn
    heat 0.2
    0 / 1 mentions
  183. 183
    tomek7667/cbjev
    Typed decisions from one encoder pass for LLM guardrails.
    GitHub · ★ 5 · Python · first seen 2026-09-24 · hn
    heat 0.2
    0 / 1 mentions
  184. 184
    nwjang/psychosis-guard
    Trajectory-aware guardrails catching gradual delusion reinforcement in LLM chatbots.
    GitHub · ★ 4 · Python · first seen 2026-09-23 · hn
    heat 0.2
    0 / 1 mentions
  185. 185
    kapadias/nonna
    Test suite runner and git hooks blocking unsafe agent actions.
    GitHub · Shell · first seen 2026-10-01 · reddit
    heat 0.2
    0 / 1 mentions
  186. 186
    pryxor/pryxor-demo
    Sandbox demonstrating real-time interception of AI agent actions.
    GitHub · ★ 1 · Shell · first seen 2026-10-01 · reddit
    heat 0.2
    0 / 1 mentions
  187. 187
    synentra/synentra
    Intent-aware runtime governance for autonomous AI agents interacting with APIs.
    GitHub · ★ 4 · C# · first seen 2026-10-01 · reddit
    heat 0.2
    0 / 1 mentions
  188. 188
    gregwebs/agent-vm
    Run AI agents in safe VMs scoped to local folders
    GitHub · ★ 2 · Rust · first seen 2026-09-22 · hn
    heat 0.2
    0 / 2 mentions
  189. 189
    denisrigsby/aetheria
    Windows-local fail-closed supervisor for long-running AI work loops
    GitHub · ★ 4 · Python · first seen 2026-09-22 · hn
    heat 0.1
    0 / 1 mentions
  190. 190
    discobox-ai/discobox
    Sandboxed execution environments for coding agents
    GitHub · ★ 104 · Go · first seen 2026-09-22 · hn
    heat 0.1
    0 / 1 mentions
  191. 191
    scopetrail/scopetrail
    Cryptographically signed delegation receipts for AI agents acting for humans.
    GitHub · ★ 1 · TypeScript · first seen 2026-09-22 · hn
    heat 0.1
    0 / 1 mentions
  192. 192
    agent-chaperone/agent-chaperone
    MCP proxy screening AI agent tool calls and results before execution.
    GitHub · ★ 22 · TypeScript · first seen 2026-09-21 · hn
    heat 0.1
    0 / 1 mentions
  193. 193
    generallymatthew/factlabel
    Nutrition label for AI content: audits claims, blocks errors, shows evidence.
    GitHub · ★ 1 · Python · first seen 2026-09-21 · hn
    heat 0.1
    0 / 1 mentions
  194. 194
    goulinkh/omp-semantic-policy
    Semantic policy enforcement for coding agents with model-backed evaluation.
    GitHub · ★ 1 · TypeScript · first seen 2026-09-21 · hn
    heat 0.1
    0 / 1 mentions
  195. 195
    lukstei/slop-grader
    Jev-powered grader that runs every rule against every text line.
    GitHub · ★ 37 · TypeScript · first seen 2026-09-21 · hn
    heat 0.1
    0 / 1 mentions
  196. 196
    polarisbuiltinc-wq/ora-grounding
    Post-response grounding check and adversarial review for LLM chat agents.
    GitHub · ★ 3 · Python · first seen 2026-09-21 · hn
    heat 0.1
    0 / 1 mentions
  197. 197
    doronp/jevc
    Compiles agent policy prose into deterministic verdict programs with code-based decisions.
    GitHub · ★ 8 · TypeScript · first seen 2026-09-20 · hn
    heat 0.1
    0 / 1 mentions
  198. 198
    emetgate/emetgate
    Verification kernel between LLM agents and source tree for deterministic proposal validation.
    GitHub · ★ 4 · Zig · first seen 2026-09-20 · hn
    heat 0.1
    0 / 1 mentions
  199. 199
    ghuntley/preflight
    Local proxy scanning LLM requests and attachments for secrets before model submission.
    GitHub · ★ 122 · Rust · first seen 2026-09-20 · hn
    heat 0.1
    0 / 1 mentions
  200. 200
    getjump/airbag
    Sandbox for coding agents with copy-on-write workspace and review outbox.
    GitHub · Go · first seen 2026-10-12 · hn
    heat 0.1
    0 / 1 mentions