ClawHub
Back to skill

Security audit

Agorism in the 21st Century Vol. 1

Security checks for vulnerabilities and agentic risk

Overview

This is a static Markdown knowledge-base skill about agorist philosophy; it contains politically and legally sensitive examples but no hidden code, persistence, credential access, or automatic actions.

Review this skill as politically and legally sensitive reference material. It should not be treated as advice to evade taxes, smuggle goods, build dark markets, bypass regulation, or hide unlawful activity; users should keep requests within applicable law and policy.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The content explicitly frames illegal acts such as tax evasion, smuggling, illegal distribution, darknet-style trading, and unregulated DeFi activity as examples of agorist practice, and it does so in a normalizing or promotional tone without legal or safety warnings. In a skill intended as a reusable knowledge base, this can encourage users to operationalize unlawful conduct or financial-risk behavior, especially because the material links ideological framing to concrete tactics and modern crypto infrastructure.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The entire skill file is written only in English and does not offer any language or locale choice or opt-in. Under the stated policy, language-policy issues are the only natural-language policy concern in scope for all file types, so this file's fixed monolingual presentation may be considered a locale/language constraint.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
65% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · ch05-pandemos.md (reported line 11)May include surrounding context.

md
- **Empire → Church → Reformation (the master analogy)**: after Rome's physical power died, the Catholic Church held "an empire of the mind" for a thousand years until printing triggered Reformation/Enlightenment. Parallel: after the American empire, the Internet/protocols remain as the church of psychological order; cryptocurrency as potential Reformation — putting "the power of computer programming into the hands of the people" as reformers did reading and writing.
- **Cybernetic genealogy of the Internet**: Wiener's cybernetics ("control and communication in the animal and the machine," homeostasis, κυβερνητική = "that which belongs to governance") — Wiener himself warned AGAINST applying it to humans; post-WWII technocrats applied it anyway (democracy needed reinforcement against fascism). Licklider/Taylor's ARPANET = "computers as a communication device" — cybernetic social feedback finally given technological foundation.
- **Pandemonium architecture (Selfridge 1959)**: many small demons each doing limited recognition tasks, feeding results upward — "the entire pandemonium could understand and make complex decisions." Today's AI still works this way: cognitively opaque, "a sort of non-human practical knowledge," fed by Internet-converted human activity, modulating society via cybernetic feedback.
- **Societies of control → hypercontrol**: Deleuze's "Postscript" (codes/passwords, "dividuals" — individuals become markets, samples, data) extended by Agamben (per Halpin, Venice seminar): hypercontrol = all social life subsumed into digital totality, pandemic-digitized even in resistant institutions. Stiegler: the "shock of the digital" short-circuits individuation, fractures inter-generational transmission, closes the horizon of the future ("a metaphysical halting of the future" — Heidegger's calculation become universal; Schürmann: cybernetic homeostasis forever, no binding principle left).
- **Techniques vs technology (Invisible Committee via Agamben)**: techniqu
...[truncated 26 chars]

Scope Creep

Low
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · ch05-pandemos.md (reported line 39)May include surrounding context.

md
2. The Internet is the church that outlives the American empire; crypto could be its Reformation.
3. Cybernetics was social control from the start — Wiener warned; ARPA built.
4. Hypercontrol = control-society + digital totality; dividuals are its subjects.
5. Techniques without technology: extend capabilities without feeding the control system.
6. Crypto-anarchy's goal: a world without guiding principle (an-archy) — destiny in every people's own hands.

## Connects To

Static analysis

No suspicious patterns detected.