extension
@grothesque/sandburg
Pi sandbox balancing security, usability and agent autonomy; for Linux
+3
Packages
6.2k listings for pi
🛡 Guards & policy Block or rewrite dangerous tool calls, sandbox commands, enforce permissions.
323 results · page 2 of 7
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.
skill
choose-security
Install the Choose-Security 3-skill security chain (review → document → remediate) into Claude Code, pi, and opencode in one command.
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.
extension
@dihak/pix-gate
Pi extension — permission gate for dangerous bash commands (confirm/block with TUI dialog)
extension
@kjrjay/pi-sandbox
Run Pi file and shell tools in an isolated Apple Container or Docker workspace
+4
extension
pi-opa
Enterprise-grade Open Policy Agent (OPA) integration for pi coding agent - comprehensive authorization, authentication, and policy enforcement
+2
extension
@simplecyon/pi-safe-operation
Operation safety and secret egress guardrails for Pi
+4
tool
@tangle-network/tcloud-agent
Agent run-loop primitive over Tangle sandbox transports — runs an AgentProfile against a brief with criterion gates, budget caps, and streaming events. Includes TangleToolProvider for Pi tool integration.
+1
extension
pi-shepherd
Line count guard and behavior rules extension for pi-coding-agent
+4
extension
@giuseppe.trisciuoglio/pi-prevent-destructive-commands
Unconditionally blocks destructive bash commands before execution. A pi extension that guards against rm, git reset --hard, docker rm, aws delete operations, and more.
promptextension
@dougbots/pi-agents
Agent profiles for pi — preconfigured model, system prompt, tool restrictions, and permission gates
+1
toolextension
pi-perms
Pi CLI extension — configurable permission policy from .agents/permissions.json and native agent configs
extension
@akshaykarle/pi-tools
Pi coding agent extensions — security hardening, agent teams and more
+7
toolextension
pi-agent-permissions
Pi CLI extension — configurable permission policy from .agents/permissions.json and native agent configs
promptextension
@hank-warren/pi-permission-selector
Numbered approval options and Tab-to-comment for Pi's extension select dialogs, including pi-auto-permissions command approval prompts.
extension
pi-permissive
A deliberately permissive permission gate for the Pi coding agent — blocks only destructive commands, never nags.
extension
@komarspn/pi-permission-system
Permission enforcement extension for the Pi coding agent.
+2
extension
pi-armory
Declarative command tools for pi — structured, pre-approved commands instead of a shell.
+4
skill
pi-dependency-safety
Pi skill for reviewing npm-registry dependency safety across npm, pnpm, yarn, and bun before project initialization, feature work, package additions, or package updates.
extension
@4meta5/pi-ozcar
Pi-first audit extension package for structured security reviews and deterministic comparison exports
extension
pi-autonomy-profiles
Standalone Auto Mode commands, permission modes, and guardrails for Pi
extension
pi-agent-sandbox
Sandbox awareness for pi: shows a colored [sandboxed:<profile>] footer and injects sandbox-access context when pi runs inside agent-sandbox (asb), plus an asb-pi shell wrapper that launches pi sandboxed inside tmux.
extension
pi-focus-guard
Pi extension that groups discuss, write, and commit guards under focus-prefixed commands.
extension
@uiyzzi/pi-asroot
sudo for pi — agent asks, user types their password into a masked TUI input, password never reaches the model
skill
@getpipher/quality
Code quality skills for pi — linting/fixing, TypeScript strict checks, production-readiness audit, multi-persona QA, and a brutal pre-prod code roast.
extension
@bytesbrains/pi-agent-supervisor
Runtime safety net for AI agents — blocks dangerous commands, protects files, enforces rate limits, and records sessions.
skillextension
@haemmid/pi-skill-paths
Canonical SKILL.md path correction and bash skill-access guard for Pi.
skillextension
pi-chefs
Long-running expert Pi sessions (chefs) that other agents consult via pi-postman. Persona + skill-allowlist layer on top of pi-postman.
+2
extension
@yoseph_23/implannotator
Approval-gated frontend implementation for Pi, combining Implannotator design guidance with Plannotator reviews.
+3
extension
@estebanforge/pi-js-review
JavaScript code-review tool for Pi. Grades your git diffs against a focused JS flaws rubric (5 sections, 18 entries: equality/coercion, async, mutation/scope, globals/prototypes, security). Each entry has a bad/good pair; the LLM proposes a corrected snip
extension
@sycghj/pi-permission-system
Permission enforcement extension for the Pi coding agent.
+4
extension
@baryonlabs/pi-dynamic-workflows
Claude-Code-style dynamic workflow orchestration for Pi (Baryon fork: crash-safe script errors + guarded sandbox).
skillprompt
@amitkot/pi-permission-tune
Pi skill for maintaining @gotgenes/pi-permission-system config. Covers pattern matching, safe vs dangerous commands, and the prompt-to-rule workflow.
extension
pi-switch-model
Free-form, LLM-driven model switching for Pi — resolve a free-text model reference and switch mid-conversation, no approval loop, no auto-restore
extension
pi-command-guardian
Deterministic explicit deletion guard for Pi
extension
pi-pip2uv
A pi extension that blocks pip install-style Python dependency changes and guides agents to use uv.
extension
@ikuradon/pi-guardian
Codex-style sandbox boundary guardian gate for pi-coding-agent.
+1
promptextension
@jaraxxxx/pi-extensions
A comprehensive collection of extension packages for Pi the coding agent — token guard, system context, tool profiler, project detector, dirty guard, permission gate, prompt customizer, auto commit, turn clock, branch navigator, and more.
+1
extension
@oracleot-tools/safe-install
Pi slash-command package that audits npm packages before installing them with --ignore-scripts.
toolextension
cc-safety-net
A coding agent CLI hook - block destructive git and filesystem commands before execution
+1
extension
betterwright
A persistent, policy-guarded Playwright browser for AI agents with network controls, trusted credential filling, proof screenshots, and CAPTCHA helpers.
+4
skillprompt
@piagent/platform
Reusable Pi agent platform, project adapters, MCP/tool policy, and coding workflow harness.
+4
extension
@spences10/pi-redact
Tool-output redaction for Pi that replaces likely secrets before they reach the model context
extension
@aliou/pi-guardrails
+2