promptextension
picodesandbox
OS-level sandboxing for pi with interactive permission prompts
+1
Packages
6.2k listings for pi
🛡 Guards & policy Block or rewrite dangerous tool calls, sandbox commands, enforce permissions.
286 results · page 1 of 6
extension
@spences10/pi-nopeek
Secret-safe workflow reminder that steers Pi agents to use nopeek instead of exposing .env values
extension
pi-mono-sentinel
Pi extension that guards against content-based secret leaks and indirect script execution
extension
pi-mono-context-guard
Pi extension that guards context window growth by auto-limiting read and rg output
+1
promptextension
@diegopetrucci/pi-permission-gate
A pi extension that prompts before dangerous bash commands and protected file writes.
extension
pi-agentkernel
Route Pi coding agent commands through agentkernel microVM sandboxes
+2
toolextension
@melihmucuk/leash
Security guardrails for AI coding agents
+1
extension
@upstash/box-pi
Pi coding agent extension that runs all tool calls inside a remote Upstash Box sandbox while the agent runs locally
+2
toolextension
@boozedog/pi-codemode
Pi Codemode plugin - TypeScript code execution with sandboxed tools, just-bash shell, and MCP integration
+3
extension
@nightona-co/pi
Pi coding agent extension that runs all tool calls inside a remote, ephemeral Nightona sandbox while the agent runs locally
+2
extension
pi-show-diffs
Pi package that adds a diff approval viewer before edit and write tools change files.
+1
extension
@yevhen.b/pi-preflight
Tool-call approvals and policy rules for Pi.
+2
skilltool
pi-daytona
A pi extension that runs tools inside a Daytona cloud sandbox for safe, isolated execution. v0.0.2 — per-session sandbox state supports concurrent sessions; skill files read locally first.
+2
extension
pi-thegreataxios-staples
Personal staple extension for pi coding-agent — protected paths
extension
pi-protected-paths
Blocks write/edit/read of sensitive files (.env, .git, node_modules, SSH keys, credentials, and more)
extension
@pi-lab/permissions
Permission system extension for pi coding agent
toolextension
pi-codex-security
Pi extension for OpenAI Codex Security — scan repositories for vulnerabilities, review findings, and fix them with the pi agent
toolextension
pi-casefile
Offensive security case tracker for pi and Codex — bug bounties, CTFs, security audits
+2
extension
@yassimba/pi-guardrails
Editable distribution of Pi Guardrails for local policy customization
+3
extension
pi-1password
1Password Pi coding agent secret management
+1
extension
pi-bash-readonly
Sandboxed read-only bash for Pi agents via bwrap
+1
extension
@a5c-ai/babysitter-pi
Orchestrate complex, multi-step workflows with event-sourced state management, hook-based extensibility, and human-in-the-loop approval — pi
extension
pi-deepseek-cache
DeepSeek prefix-cache extension for pi — hit-rate telemetry, prefix guard, and cache-friendly compaction.
+2
toolextension
@m4riok/pi-ide-bridge
Pi extension for VS Code diff approval workflows and editor context bridging.
+4
extension
@christianmoesl/pi-sbx
Run Pi coding-agent tool calls inside Docker sbx sandboxes
+2
extension
@liziy/plan-guard
Tab 键切换计划/执行模式
extension
pi-sensitive-guard
Pi extension that protects sensitive files, blocks secret writes, and optionally redacts protected read output.
+1
extension
@celestoai/pi
Run Pi coding tools in a Celesto computer
+4
extension
@alexleekt/pi-pkg-guard
Don't let orphaned packages clutter your agent. Guard against unregistered pi packages.
extension
@jerryan/pi-bash-wrap
Sandbox pi bash commands with bubblewrap
+1
extension
@senad-d/guardme
Deny-first Pi guardrails that keep LLM shell and file access safe, transparent, and user-approved.
extension
@thurstonsand/pi-permissions
Pi extension package for user, project, and package-level tool permission hooks
+1
extension
@senad-d/protectme
Pi extension that guards agent network and website access with global and project allow lists.
+1
extension
pi-supersafety
When the example safety extensions don't cover what you need, this should! Simple with approval flow and windows sandbox capability (with Sandboxie)
extension
@0xkobold/pi-whitelist
Tri-state tool permission system (allow/deny/ask) for AI agent tool invocations
extension
pi-gondolin-mount
pi extension that sandboxes LLM coding agents inside a Gondolin micro-VM with configurable additional mounts
+1
extension
@nklisch/pi-enhanced
Pi, enhanced — one install for nklisch's full harness: policy-gated command review, plugin marketplace, subagents, background tasks, research tools, search, model modes, and a curated UX set.
+8
skillextension
loop-skills
Research-driven planning, debugging, and audit skills for Codex, Pi, and Claude Code
+2
extension
pi-tool-wal
Write-ahead log for Pi tool calls: record intent before execution, reconcile results after, persist to SQLite for audit and crash recovery
skillextension
agent-booster-pack-skills
General Agent Booster Pack skills for Pi: data, security, debugging, refactoring, and more. Runtime-owned skills ship with their matching extension packages. Renamed from pi-agent-booster-pack; old name deprecated.
extension
pi-perm
Config-driven Pi sandbox and permission extension using sandbox-runtime.
+1
extension
pi-wayfinder-guard
Pi extension enforcing no-implementation during wayfinder (fog mode), with a parallel four-axis code review.
+2
extension
@grothesque/sandburg
Pi sandbox balancing security, usability and agent autonomy; for Linux
+3
extension
@aslamplr/pi-safe-shell
Protect your production assets from dangerous bash commands. Default no-bash mode with user approval gates and whitelist support.
+3
extension
pi-claude-permissions
Claude-style allow/deny/ask permission lists for Pi tool calls
skillextension
agent-booster-pack-contract-first
Interface Design Gate for Pi plus contract-first skill: pause and require human approval of contract/API shape before implementation lands.
extension
@aprimediet/permission-modes
A Claude-Code-style **Shift+Tab mode cycle** for the [pi coding agent](https://pi.dev).
+2
extension
pi-refusal-guard
A Claude safety-classifier refusal shouldn't kill your turn — rescue, retarget the server-side fallback chain, and see what's tripping.