claudecodex

proteus

A plugin for Claude Code and Codex that coordinates structured, continuous vulnerability research with memory, strategy, and realistic validation.

35 on GitHubby Vyntra Researchv2.1.11GPL-3.0-or-laterupdated 26 days agoDeveloper Tools
/

Plugin proteus from marketplace proteus-marketplace in Vyntra-Research/Proteus. Commands read from the repo's manifest — 26 days ago.

This plugin runs code on your machine — it starts local MCP servers. The exact commands are listed below; read them before installing.

What it installs

Skills (10)

  • chaining

    Develop non-obvious exploit chains from concrete primitives, side effects, trust-boundary drift, state transitions, component coupling, and capability amplification. Use when a Proteus branch needs creative offensive reasoning beyond the obvious direct bug path.

  • checkpoint

    Compress Proteus campaign or round state after meaningful progress, preserving confirmed facts, killed paths, pivots, branch scores, learned heuristics, unresolved blockers, and the next high-ROI move.

  • chimera-agent

    Operate as a secondary OpenCode-backed Proteus Chimera agent with coordinator-controlled scope, communication, snapshots, labs, and permissions.

  • codebase-research

    Perform deep offensive codebase research for Proteus by mapping architecture, dataflow, trust boundaries, invariants, side effects, recent change risk, and high-ROI branch material. Use when Codex must understand a codebase before chaining, fuzzing, Cicada, or PoC work.

  • continuous-vuln-research

    Coordinate disciplined, continuous, professional vulnerability research with Proteus memory, campaigns, bounded delegation, validation gates, realistic exploitability, anti-slop controls, and report-grade decision discipline.

  • fuzzing

    Design calibrated fuzzing, differential probes, harnesses, and oracles for narrow Proteus surfaces. Use when the goal is to learn how inputs affect parsers, state machines, adapters, runtimes, or side effects, not to spray generic payloads.

  • poc-exploit

    Build realistic local/OSS-authorized PoC plans, labs, manual blackbox reproduction steps, negative controls, and exploitability evidence for Proteus candidates with concrete signal.

  • post-ai-blind-spots

    Close shared-premise blind spots around an established real sink by mapping every reachable natural path and CIA outcome before kill, pivot, exhaustion, or safety claims. Use when behavior looks correct or intentional, implementation and validation may repeat one premise, or a chain crosses componen

  • web-intel

    Gather security intelligence for Proteus hypotheses: expected behavior, public-known status, advisories, changelogs, issues, PRs, docs, tests, affected-version timeline, duplicate risk, and research pivots. Use when a branch needs external truth, not active web exploitation.

  • web-research

    Conduct authorized web application research for Proteus targets using the same campaign, chaining, fuzzing, intel, and PoC heuristics adapted to live or local web surfaces. Use for endpoint behavior mapping, blackbox probes, browser/API workflows, and web pentest-style validation.

Slash commands (1)

  • /proteus:proteus

    Run Proteus continuous vulnerability research for the current target.

Subagents (8)

  • proteus-argus

    MUST BE USED for Proteus component-level review fronts where a bounded module, component, or primitive needs detailed security inspection.

  • proteus-artificer

    MUST BE USED after initial Proteus gates pass to build realistic PoC labs, manual repro steps, negative controls, and triage-ready report drafts.

  • proteus-chaos

    MUST BE USED for Proteus fuzzing, edge-case, parser, canonicalization, cache-key, or anomaly-matrix fronts.

  • proteus-cicada

    MUST BE USED only when a Proteus branch already has concrete signal and needs exploit-development, bypass, chaining, reliability, or impact-proof work.

  • proteus-libris

    MUST BE USED before any Proteus report-grade claim to verify docs, tests, contracts, advisories, issues, changelogs, public-known behavior, and timeline.

  • proteus-loom

    MUST BE USED for Proteus macro or chaining analysis where separate primitives may combine into a realistic exploit path.

  • proteus-mimic

    MUST BE USED for Proteus runtime, adapter, deployment, build, Docker, WSL, native, or environment-divergence fronts.

  • proteus-skeptic

    MUST BE USED before any Proteus report-grade claim to refute, downgrade, or kill candidates through adversarial review.

MCP servers (1)

  • proteus

    runs: node ${CLAUDE_PLUGIN_ROOT}/scripts/proteus-mcp.cjs

View source on GitHub

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

Scaffold, validate, and publish new Claude Code plugins with proper structure

3 skills · 1 command · 1 agent

ruvnet/ruflo 81,589

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

slidev

claude

Claude Code skill for creating and presenting Slidev slidedecks

1 skill

slidevjs/slidev 49,452

gitnexus

claudecodex

Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.

12 skills · 1 MCP server · hooks

abhigyanpatwari/GitNexus 48,482

Maintain proteus?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted