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

@wiolett/workflow

v0.2.7

Published

Workflow MCP runtime for custom-agent sync plus filesystem-backed plan, audit, handoff, and UI contract artifacts.

Readme

Workflow

MCP runtime for the Workflow Codex plugin.

On server startup it validates the packaged workflow_*.toml custom-agent definitions and syncs them globally into ~/.codex/agents/. Canonical agents route clear mechanical work to GPT-5.6 Luna, everyday implementation/review to GPT-5.6 Terra, and complex/high-assurance work to GPT-5.6 Sol; skills reference semantic work classes rather than model-version strings.

It also creates best-effort compatibility links in ~/.agents/agents/ that point to the synced Codex agent files. If symlinks are unavailable, it falls back to managed copies. Locally modified or unmanaged compatibility files are not overwritten.

Agent sync is a startup invariant, not a model-visible action, so the server does not expose sync tools.

The Workflow plugin is also the consolidated hook owner for Wiolett plugins. Its hook can detect sibling agent-memory and merge-request-review plugin installs and add the relevant startup context or merge_request_* reviewer validation without those plugins registering separate hooks. A Codex-only Stop adapter enforces pending material-plan reflection; Claude Code remains hook-optional because the portable behavior lives in skills and MCP state.

Tools

The MCP tools are deterministic filesystem helpers for workflow artifacts, stored under .workflow/ by default. Use them whenever available for workflow status, run creation, state updates, run completion, artifact writes, findings normalization, structured handoffs, and bounded same-model commitment reflection; manual writes are fallback only. They do not generate plans, run agents, or make architecture decisions.

Workflow reads mcp.workflow.artifacts from $AGENTS_HOME/.wiolett/config/mcp-config.yml. It never creates or modifies the file; when the file or section is absent it uses .workflow, plans, and audits. Invalid configuration produces a warning and the same defaults.

State update tools reject unknown operations with a nearest supported operation and a payload hint. Agents should correct the operation shape and retry the MCP call instead of falling back to manual .workflow/ writes. Plan and audit updates use one shared operation handler, so the documented plan/audit operation lists are the meaningful subsets agents should use for those run types rather than separate runtime validators.

Plan state tracks active_chunk and chunks[]. Use set_active_chunk, clear_active_chunk, complete_chunk, cancel_chunk, and wait_chunk for chunk lifecycle; use upsert_chunk for chunk metadata.

  • workflow_status: read active plan/audit state and latest workflow runs.
  • workflow_plan_create: create .workflow/plans/<slug>/ with plan.md, manifest.json, state.json, context/questions/decisions files, ui-contract.md, artifacts/, chunks/, and handoffs/.
  • workflow_plan_update: apply structured state operations to the active or named plan run.
  • workflow_plan_commitment_propose: record a material candidate, detect scope pressure, and return one shrink-first reflection prompt.
  • workflow_plan_commitment_confirm: record KEEP, SHRINK, ASK, or REPLAN; only a reviewed material commitment may execute or complete.
  • workflow_plan_complete: mark the active or named plan complete and clear active_plan when it points to that run.
  • workflow_plan_artifact_write: write allowed plan-run files without path escape.
  • workflow_audit_create: create .workflow/audits/<slug>/ with audit state, prompt/review/sanity folders, findings, and master audit files.
  • workflow_audit_update: apply structured state operations to the active or named audit run.
  • workflow_audit_complete: mark the active or named audit complete and clear active_audit when it points to that run.
  • workflow_audit_artifact_write: write allowed audit-run files without path escape.
  • workflow_handoff_write: write structured module handoffs into active plan/audit state.
  • workflow_findings_normalize: normalize findings into stable severity-sorted workflow findings.

If an installed MCP launched through @wiolett/workflow@latest exposes fewer tools than this source list, treat it as package/install drift and check the published package version before changing the source contract.