npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@oppiai/pi-package

v0.2.11

Published

OPPi Pi package: extensions, skills, prompts, and themes.

Readme

@oppiai/pi-package

OPPi's Pi package: extensions, skills, prompts, and themes that make stock Pi feel like OPPi.

Most users should install the CLI instead:

npm install -g @oppiai/cli
oppi doctor
oppi

For direct Pi debugging or package development:

pi --no-extensions -e ./packages/pi-package

The package registers OPPi defaults, model-facing feature routing guidance, docked UI, /prompt-variant A/B prompt selection, /effort, /permissions, /review, /init, /agents, /runtime-loop, /independent @plan.md, /exit, /clear//reset, todo_write, ask_user, suggest_next_message, shell_exec/shell_task, feedback intake, image_gen, render_mermaid, themes, terminal setup, configurable usage/footer bars, follow-up chain context, Hoppi memory hooks, explicit @oppiai/hoppi-memory install prompts/settings, legacy /memory-maintenance fallback cleanup (superseded by automatic dreaming when enabled), and compaction helpers.

/agents browses built-in personalities (general-purpose, Explore, Plan, oppi-code-guide, statusline-setup, verification) and manages OPPi agent definition markdown files in .oppi/agents and the personal OPPi agent dir, while also listing compatible .claude/agents files as read-only imports. Project/personal agents can override built-ins by name. It manages definitions only; full AgentTool runtime dispatch belongs to the Rust runtime spine.

The Rust-loop bridge now defaults to default-with-fallback: normal Pi turns keep using Pi as the primary runtime while redacted lifecycle events are mirrored into oppi-server and finalized through Rust's 11-phase turn/run-agentic path when the server is available. /runtime-loop <prompt> still forces a single mirrored turn, /runtime-loop status reports the last capture, and /runtime-loop cancel stops the bridge capture while leaving Pi's stable runtime fallback alone. Set OPPI_RUNTIME_LOOP_MODE=command for opt-in-only mirroring or off to disable the bridge. Separately, the CLI-level oppi runtime-worker smoke --json exercises the direct Rust provider worker path with a local OpenAI-compatible mock, including streaming chunks, compact transcript seeding, a provider-requested Rust tool call, Rust-owned text file tools, and approval-resume support. oppi runtime-worker <prompt> is now the opt-in direct-worker dogfood command for credentialed provider runs; when Hoppi is installed and memory is enabled, the CLI bridges bounded Hoppi recall into the direct-provider context and saves a bounded Hoppi turn summary after completion. The direct worker now also honors the same prompt A/B variant setting (--prompt-variant, OPPI_SYSTEM_PROMPT_VARIANT, stored oppi.promptVariant), Rust owns oppi__todo_write state with todos/list for the backend half of /todos parity, and Rust applies host-supplied follow-up chain context to direct-provider prompts while Pi/native clients own the visible queue UI. It reports stable Pi fallback guidance when provider auth or oppi-server is unavailable, and supports --no-memory / OPPI_RUNTIME_WORKER_MEMORY=off for memory-free dogfood. Precise tool schemas, a generalized Rust memory broker for non-CLI clients, native /todos UI, native follow-up queue UI, and persisted provider snapshots remain future direct-worker slices.

/independent @plan.md starts a reusable plan-runner workflow: it loads the independent skill, reads the referenced plan document, maintains todos, asks only blocking clarification questions, and continues through implementation plus validation instead of stopping at a proposal.

shell_exec gives the model a bounded Bash/PowerShell execution tool with cwd tracking, timeouts, output logs under output/shelltool, and optional background execution. Use shell_task to list, read, or kill background shell tasks.