@subaud-tools/pi-permission-system
Permission enforcement extension for the Pi coding agent.
$ pi install npm:@subaud-tools/pi-permission-systemPackage Catalog
Extensions, skills, prompt templates, and themes published to npm. Install with pi install npm:<package>. See the package docs for details.
Recently published
@rsrsai/piNative Respire memory extension for Pi coding agent4m ago@llblab/pi-kitVersion-pinned distribution of LLB Lab extensions and Skills for Pi6m agoshariq-pi-extensionsCross-platform extension suite for the Pi coding agent.6m ago@piagent/platformGuardrails and managed team harness for AI coding agents, with isolated tools, scoped approvals, and a local dashboard.20m ago@wezzard/mcp-relayMCP server and Claude Code plugin. Runs an agent's code on fresh, isolated VMs, several at once, each run recorded with evidence20m ago@a-t-h-i/bot-lobbyStructured multi-agent software engineering orchestrator for Pi20m agopi-lensReal-time code feedback for pi — LSP, linters, formatters, type-checking, structural analysis & booboo20m agoavtc-pi-featyardPredictable, deterministic feature development for pi — deep upfront design, configurable comprehensive review and verification rigor, and auto-agents draining a backlog22m ago
All packages
51-87 / 87 (of 5380)
@meowshed/meow-unattended
Prints the posture an unattended run is held to from the [unattended] table a repository declares: the permission mode, the gates the run may decide, its release command and whether it may amend an approved record, with each deny rule and what the rules d
$ pi install npm:@meowshed/meow-unattended@gotgenes/pi-permission-model-judge
Deny-first typo-path model judge — a pi-permission-system Authorizer chain link
$ pi install npm:@gotgenes/pi-permission-model-judgepi-jev-auto-mode
Jev (TypeSafe System One) backed auto mode for the Pi coding agent: semantically auto-approves bash, write, and edit tool calls and fails closed when a decision cannot be made.
$ pi install npm:pi-jev-auto-mode@komarspn/pi-permission-system
Permission enforcement extension for the Pi coding agent.
$ pi install npm:@komarspn/pi-permission-system@pify/yolo
A safety gradient from auto-approve-everything to ask-about-anything: four modes, read-before-write, undo trail, rewind to before a prompt
$ pi install npm:@pify/yolopi-amplike
Pi skills and extensions that provide Amp Code-like workflows (handoff, permissions, mode selector, web access).
$ pi install npm:pi-amplike@valdo766hi/pi-yolo
Pi extension providing a session-local permission approval overlay
$ pi install npm:@valdo766hi/pi-yolopi-governance
A pi extension that turns every pi tool call into a permission request to the ACP host before it runs.
$ pi install npm:pi-governancepi-governor
A guard layer for pi tool calls: rules that allow, ask or never allow each call, shared by every package
$ pi install npm:pi-governorpi-permission-layers
Pi extension to handle permissions using a layered approach
$ pi install npm:pi-permission-layers@pi-lab/permissions
Permission system extension for pi coding agent
$ pi install npm:@pi-lab/permissionspi-permission-explain
On-demand plain-English explanations for pi-permission-system prompts
$ pi install npm:pi-permission-explainpi-permission-classifier
Auto-classifier Authorizer chain link for pi-permission-system: a light model reviews each ask and returns allow, deny, or defer
$ pi install npm:pi-permission-classifier@aemonge-dev/pi-openspec-gate
Declared-work gate for Pi: accounting at the tool-call boundary — every mutation is exempt-recorded, attributed to a focused change, declared by the human, or blocked with remedies. Never permission; never a renderer; never a prompt. Flow vocabulary is sc
$ pi install npm:@aemonge-dev/pi-openspec-gateone-code-extension
One Code as a pi package — the Claude Code experience on your own pi install: ported system prompt, Claude-Code-format permissions, .claude/ compatibility, ToolSearch-style deferred tools, and system-reminder steering. For the bundled app, install `@one-a
$ pi install npm:one-code-extension@bacnh85/pi-permission
Granular permission system for Pi — allow/ask/deny rules per tool with wildcard patterns, external-directory boundary, and a doom-loop guard. Config-driven, opt-in companion to Pi's container-first philosophy. Inspired by OpenCode's permission config.
$ pi install npm:@bacnh85/pi-permissionpi-async-tools
One async tool for pi: run any callable tool in the background and get its result later as a message, with the same hooks, sandbox and permission checks.
$ pi install npm:pi-async-tools@riposta/pi-jev
A Jev classification layer for the Pi coding agent: typed routing, permission gating, context shielding, pruning and loop detection.
$ pi install npm:@riposta/pi-jev@restranger/repi-permission-system-modes
Permission modes (default, bypass, deny) and /mode command for pi-coding-agent
$ pi install npm:@restranger/repi-permission-system-modes@shelken/pi-guard
Hard-block dangerous agent bash commands and secret path access
$ pi install npm:@shelken/pi-guard@bryan2333/pi-permission-system
Permission enforcement extension for the Pi coding agent.
$ pi install npm:@bryan2333/pi-permission-systempi-workflows-claude
Claude-Code-style workflow engine for pi: a Workflow tool that runs deterministic JavaScript orchestration scripts (agent/parallel/pipeline/phase) with resume, journaling, phases, permissions and a live web UI.
$ pi install npm:pi-workflows-claude@dougbots/pi-agents
Agent profiles for pi — preconfigured model, system prompt, tool restrictions, and permission gates
$ pi install npm:@dougbots/pi-agents@aefree/pi-safety-rails
Pi safety-focused extensions for file-tool path permissions and tool result hygiene.
$ pi install npm:@aefree/pi-safety-railspi-file-permissions
Pi extension that enforces file-level permissions via a YAML config. Controls which paths each tool (read, write, edit, find, grep, ls) can access.
$ pi install npm:pi-file-permissions@finngi/pi-permissions
Approve Pi tool calls by rule, by asking you, or with a classifier model
$ pi install npm:@finngi/pi-permissions@georgedong32/permission-modes
Claude-Code-style permission modes (ask / plan / auto) for the pi coding agent — with per-mode model profiles, outside-cwd write tracking, and per-mode skill filtering.
$ pi install npm:@georgedong32/permission-modes@pithos-kit/aegis
Protect Pi agent shell commands and file mutations with configurable block and confirmation rules.
$ pi install npm:@pithos-kit/aegispi-auto-approval
AI classifier based automatic approval extension for Pi coding agent.
$ pi install npm:pi-auto-approvalpi-muselinn-harness
Kimi Code-style agent orchestration harness for Pi coding agent — Swarm (concurrent subagents + braille TUI), Goal (lifecycle + budget + queue), Plan (plan mode + tool restrictions), Permission (18-level policy chain: auto/yolo/manual), Task (background +
$ pi install npm:pi-muselinn-harness@wengine-ai/pi-session-handoff
Hand a self-contained task to another local pi session (GUI or TUI) — recipient keeps its own project, model and permissions; busy targets queue instead of being interrupted.
$ pi install npm:@wengine-ai/pi-session-handoffpi-permission-guardian
Permission guardrail extension for the pi coding agent.
$ pi install npm:pi-permission-guardian@foxfirecodes/pi-sandbox
OS-level sandboxing for pi with interactive permission prompts (fork)
$ pi install npm:@foxfirecodes/pi-sandbox@mattschoe/pi-plan
Plan mode, accept-edits mode, readable bash approval reasons, and plan execution tracking for Pi.
$ pi install npm:@mattschoe/pi-plan