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

taintctl

v0.0.0

Published

Content-aware provenance layer for Claude Agent SDK. Detects dangerous content at every tool I/O boundary, propagates classification across sub-agent dispatch, and enforces fail-closed policy.

Downloads

137

Vulnerabilities

Powered byPowered by Snyk
  • 0High
  • 0Medium
  • 0Low

Links

Git

Maintainers

jokfajokfa

Keywords

ai-securityagent-securityclaude-agent-sdkguardrailstaint-analysisprompt-injectionmcpprovenance

Readme

taintctl

Content-aware provenance layer for Claude Agent SDK and other agent frameworks. Detects dangerous content at every tool I/O boundary, propagates classification across sub-agent dispatch, and enforces policy with a fail-closed default.

Status: pre-alpha. Design phase. No runnable code yet.

Why this exists

Agent systems built on Claude Agent SDK (and similar orchestrator-worker patterns like LangGraph, CrewAI) recognize dangerous content unevenly across calls. When agent A reads .env and dispatches a sub-agent with that data in its prompt, sub-agent B has no signal that the data was already classified as sensitive.

Existing guardrails operate at single-LLM-call granularity and do not propagate classification state across sub-agent dispatch boundaries. Existing MCP scanners operate on static descriptions and trust-on-first-use, not runtime data flow.

taintctl fills that gap.

What's different

| Tool | What it does | Cross-subagent provenance? | Live visualization? | |---|---|:-:|:-:| | mcp-scan | Static MCP description scanning | ❌ | ❌ | | mcp-context-protector | Trust-on-first-use config pinning | ❌ | ❌ | | Lakera Guard / NeMo / guardrails-ai | Single-call content classification | ❌ | ❌ | | Claude Code permission system | Syntactic allow/deny prompts | ❌ | ❌ | | taintctl | Content classification + cross-subagent ledger + flow graph UI | ✅ | ✅ (Stage 3) |

Roadmap

| Stage | Weeks | Deliverable | |---|---|---| | 0 (Pre-code) | Week 0 | Verify SDK hook coverage, name reserved, baseline benchmarks captured | | 1 | Weeks 1-4 | Claude Agent SDK middleware + content classifiers + policy engine + terminal UI | | 2 | Weeks 5-7 | Cross-subagent provenance (channel-a fingerprint + channel-b prompt injection) | | 3 | Weeks 8-12 | Static SPA flow-graph UI + README screencast |

Full design: docs/design-2026-05-20.md Active tasks: TODO.md

Limitations (acknowledged, not hidden)

  • v1 only handles verbatim taint flow. When an LLM paraphrases sensitive data, channel-a (sha256 fingerprint) breaks. Channel-b (system-prompt warnings to sub-agents) is a partial mitigation but its effectiveness is an empirical question, not a guarantee.
  • v1 only ships a TypeScript adapter for Claude Agent SDK. Python adapter, LangGraph, AutoGen, CrewAI, OpenClaw, Hermes are v1.1+.
  • v1 prompt-injection detection is pattern-based. Paraphrased prompt injections will be missed. Documented as known gap, not silently broken.
  • Not a defense against a malicious parent agent. Standard guardrail assumption: the agent we sit inside is honest-but-naive, not adversary-controlled.

Validation

  • AgentDojo prompt-injection-marker subset: Stage 1 gate is recall ≥ 0.65 (deterministic detector floor)
  • InjecAgent: baseline numbers in CI on every PR
  • Multi-agent scenarios: 8-12 in benchmarks/multiagent/, derived from a fork of damn-vulnerable-MCP-server

License

MIT — see LICENSE

Related work

This is the author's second project in MCP/agent security. The first is MCP-Security-Framework, which scans MCP servers for vulnerable patterns. The two projects are complementary: MCP-Security-Framework is a static scanner; taintctl is a runtime provenance layer.