Sstack State Attacker — Data & AI agent for Claude Code
State lens attacker.
How to install Sstack State Attacker
Installs to ~/.claude/agents/mhenke-sstack-sstack-state-attacker.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/mhenke/sstack/HEAD/agents/sstack-state-attacker.md -o ~/.claude/agents/mhenke-sstack-sstack-state-attacker.md Restart Claude Code, or start a new session, for it to be picked up.
What Sstack State Attacker does
name: sstack-state-attacker description: "State lens attacker. Attacks every mapped surface for stale cached reads, write-through to caller data, partial updates after failure, escaped internal references, and invalid lifecycle transitions. Invoked as a subagent after the orchestrator writes the surface map. Rubric and Report format arrive inline in the dispatch."
State attacker
You are a **subagent**. The parent agent already ran Discover. Your prompt is the **user message** with la
Alternatives in Data & AI
- Docs Drift Reviewer — Checks whether clickhouse-go documentation matches changes to its ClickHouse API, database/sql API, transports 3.3k ★
- Pricing Updater — Use this agent to update LLM pricing data in the default-pricing.ts file 175 ★
- Review Code Correctness — A code review agent that checks analysis scripts for logical errors, incorrect function usage, wrong variable 134 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Sstack Ownership Attacker
Ownership lens attacker. Attacks authorization decisions across function, data, and field level: BOLA/IDOR, BO
External Data
Conditional review angle "external data" of the user's §6 gate — a feed, API or IPC boundary: reconnect, rate
Humanize Auditor
Read-only auditor for AI-voice tells in academic prose. Reviews .tex, .qmd, .md files for the 10 detection cat
Codemap Writer
Generate or refresh docs/CODEMAPS/ for the current repo. Use when the user runs /update-codemaps, when context
Architect Reviewer Complex
[COMPLEX tier (risk tier L) — opus/medium] Adversarial reviewer for a system design (design.md) or an ADR, wit
Abuse Tester
Attacks Sortd's parsers and data paths with hostile inputs (money and currency strings, dates, statement rows,
Related Skills
Build Bundle
Build one governed template bundle end to end, from research fan-out through the four-lens review to a merged
Conversation Reply
Investigate a PostHog Conversations support ticket - find root cause via git/code/PostHog data, verify whether
Agentic Usage Analysis
Skill + toolkit: audit how AI coding agents (Codex, Claude Code, OpenCode, OMP, ZCode, ...) worked on a repo —