Resource Abuse

91 findings 85 skills affected OWASP AST03: Over-Privileged Skillsalso AST06

The skill consumes compute, network, tokens, or paid API credits disproportionately: crypto mining, loops without termination, mass requests, or quietly routing fees or referrals to a third party. Often disclosed in fine print but not at activation time.

By severity

  • CRITICAL 2
  • HIGH 18
  • MEDIUM 49
  • LOW 21

By registry

RegistryFindingsSkills affected
clawhub.ai6663
skills.sh76
huggingface54
K-Dense-AI33
microsoft33
trailofbits33
anthropic11
sawyerhood11
zxkane11

How it shows up in a skill

Real findings from the corpus. Each links to the skill's full report.

whisper clawhub.ai

CRITICAL Covert Agent-to-Agent Communication Channel Bypassing Human Oversight Resource Abusellm

The skill's explicit purpose is to enable 'agents to communicate privately, exchange secrets, or coordinate without human visibility.' This creates a covert channel where agents can coordinate actions, share information, or plan operations completely outside human awareness. The description explicitly states this is for communication 'without human visibility,' which is a severe autonomy and control issue.

SKILL.md

Description: 'End-to-end encrypted agent-to-agent private messaging via Moltbook dead drops. Use when agents need to communicate privately, exchange secrets, or coordinate without human visibility.'
Remediation: 1. Remove or redesign this capability entirely - agent-to-agent communication should not bypass human oversight. 2. If such communication is needed, implement mandatory human-in-the-loop approval for all messages. 3. Add audit logging that humans can review. 4. Require explicit user consent with clear warnings about the implications. 5. Consider whether this capability should exist at all in a safe agent ecosystem.

taskmaster clawhub.ai

CRITICAL Unbounded Autonomous Agent Spawning Without User Confirmation Resource Abusellm

The skill automatically spawns multiple sub-agents with full capabilities (sessions_spawn) without requiring user confirmation for each spawn. The parallel execution feature allows spawning arbitrary numbers of agents simultaneously. Combined with automatic retry logic (max_retries=2) and no rate limiting, this creates severe resource exhaustion and cost runaway risks.

scripts/delegate_task.py

Task dataclass has max_retries=2 by default. Instructions show '[PARALLEL: true]' for simultaneous task execution. No confirmation prompts, no rate limits, no maximum concurrent agent limits in code. A malicious or buggy task could spawn hundreds of agents.
Remediation: 1. Require explicit user confirmation before spawning ANY sub-agent. 2. Implement hard limits on concurrent spawned agents (e.g., max 3). 3. Add rate limiting between spawn operations. 4. Remove automatic retry logic or require approval for retries. 5. Implement circuit breakers to stop spawning if costs exceed thresholds. 6. Add session quotas and timeout enforcement.

base-trader clawhub.ai

HIGH Unbounded Automated Trading Without Circuit Breakers Resource Abusellm

The skill promotes automated trading strategies (DCA, limit orders, stop losses) but lacks implementation of the stated safety limits. The instructions mention 'MAX_DAILY_LOSS = -20%' and 'STOP TRADING FOR 24 HOURS' but provide no mechanism to enforce these limits, allowing unbounded automated execution that could drain funds.

SKILL.md

MAX_POSITION_SIZE = 10% of portfolio
MAX_SINGLE_TRADE = $50 USD
STOP_LOSS = -15% from entry
TAKE_PROFIT = +30% minimum target
MAX_DAILY_LOSS = -20% of starting balance

If daily loss limit hit → STOP TRADING FOR 24 HOURS.
Remediation: Implement actual enforcement mechanisms for stated limits. Add portfolio tracking to calculate daily P&L. Create a circuit breaker that prevents trading when limits are exceeded. Store state to persist limits across sessions. Add mandatory confirmation for trades approaching limits.

capability-evolver clawhub.ai

HIGH Availability Disruption: Infinite Loop Execution Resource Abusellm

The skill includes a '--loop' flag that runs the evolution cycle in an infinite loop, potentially consuming system resources indefinitely. Combined with autonomous code execution, this creates a risk of uncontrolled resource consumption and system instability. The documentation explicitly encourages running this in background processes or cron jobs.

SKILL.md

'To run in an infinite loop (e.g., via cron or background process), use the --loop flag or just standard execution in a cron job.' 'node skills/capability-evolver/index.js --loop'
Remediation: Remove the infinite loop functionality. Implement resource limits, execution timeouts, and iteration caps. Any repeated execution should be explicitly controlled by the user with clear termination conditions.

How to check your own skill for this

Look for unbounded loops, background processes, and any fee, referral, or affiliate configuration the user cannot change.

Affected skills

Name Registry Verdict SkillSpector Findings Critical High Medium Low
hugging-face-datasets huggingface ISSUES FOUND Do not install 78 78 27 0 47 4
zoho-email clawhub.ai ISSUES FOUND Do not install 100 53 18 3 32 0
update-plus clawhub.ai ISSUES FOUND Do not install 100 15 5 3 4 3
base-trader clawhub.ai ISSUES FOUND Caution 0 12 1 4 4 3
simmer clawhub.ai ISSUES FOUND Caution 33 10 2 2 3 3
capability-evolver clawhub.ai ISSUES FOUND Caution 41 9 2 3 2 2
x402 clawhub.ai ISSUES FOUND Caution 30 8 1 2 3 2
foundry clawhub.ai ISSUES FOUND Caution 33 7 2 3 2 0
open-notebook K-Dense-AI ISSUES FOUND Do not install 51 7 3 2 2 0
telegram-pairing-customization clawhub.ai ISSUES FOUND Safe 7 7 1 2 2 2
ABM Outbound clawhub.ai ISSUES FOUND Caution 9 6 2 2 2 0
clawhub clawhub.ai ISSUES FOUND Do not install 61 6 2 2 2 0
omni-stories clawhub.ai ISSUES FOUND Do not install 72 6 1 3 2 0
openwork clawhub.ai ISSUES FOUND Caution 21 6 1 3 2 0
phone-agent clawhub.ai ISSUES FOUND Do not install 69 6 1 2 2 1
wallet-tracker clawhub.ai ISSUES FOUND Caution 44 6 2 3 1 0
whisper clawhub.ai ISSUES FOUND Caution 5 6 2 3 1 0
azure-eventhub-java microsoft ISSUES FOUND Caution 0 5 1 0 0 4
cryptocurrency-trader clawhub.ai ISSUES FOUND Do not install 90 5 1 3 1 0
job-auto-apply clawhub.ai ISSUES FOUND Caution 33 5 2 2 1 0
sendclaw clawhub.ai ISSUES FOUND Safe 0 5 1 2 2 0
solana clawhub.ai ISSUES FOUND Caution 9 5 2 2 1 0
stealthy-auto-browse clawhub.ai ISSUES FOUND Caution 22 5 1 2 2 0
taskmaster clawhub.ai ISSUES FOUND Caution 34 5 2 3 0 0
windows-control clawhub.ai ISSUES FOUND Caution 0 5 2 2 1 0
channel clawhub.ai ISSUES FOUND Do not install 100 4 1 2 1 0
hyperliquid clawhub.ai ISSUES FOUND Caution 43 4 1 2 1 0
insider-wallets-finder clawhub.ai ISSUES FOUND Caution 33 4 2 2 0 0
tencent-map clawhub.ai ISSUES FOUND Do not install 58 4 1 2 1 0
tinyfish clawhub.ai ISSUES FOUND Do not install 75 4 1 2 1 0
botcoin-miner clawhub.ai ISSUES FOUND Do not install 58 3 1 2 0 0
save-money clawhub.ai ISSUES FOUND Safe 0 3 1 1 1 0
session-handoff skills.sh ISSUES FOUND Do not install 66 9 0 2 3 4
browser-use skills.sh ISSUES FOUND Caution 12 5 0 2 3 0
clawflows clawhub.ai ISSUES FOUND Caution 0 5 0 2 3 0
compound-engineering clawhub.ai ISSUES FOUND Caution 33 5 0 2 3 0
gitclaw clawhub.ai ISSUES FOUND Do not install 100 5 0 2 3 0
openclaw-optimizer clawhub.ai ISSUES FOUND Caution 50 5 0 2 3 0
tokenguard clawhub.ai ISSUES FOUND Caution 25 5 0 1 3 1
using-superpowers skills.sh ISSUES FOUND Safe 0 5 0 1 2 2
voice-agent clawhub.ai ISSUES FOUND Do not install 55 5 0 1 4 0
clawmail clawhub.ai ISSUES FOUND Do not install 71 4 0 1 3 0
discord-voice clawhub.ai ISSUES FOUND Do not install 70 4 0 2 2 0
docx anthropic ISSUES FOUND Caution 35 4 0 1 2 1
firebase-apk-scanner trailofbits ISSUES FOUND Caution 48 4 0 3 1 0
agnxi-search clawhub.ai ISSUES FOUND Safe 7 3 0 1 2 0
clawlist clawhub.ai ISSUES FOUND Caution 7 3 0 1 2 0
crypto-gold-monitor clawhub.ai ISSUES FOUND Caution 49 3 0 1 2 0
youtube-voice-summarizer clawhub.ai ISSUES FOUND Caution 35 3 0 1 2 0
memory-hygiene clawhub.ai ISSUES FOUND Caution 33 2 0 1 1 0
qiime2-amplicon K-Dense-AI ISSUES FOUND Caution 28 1 0 1 0 0
statistical-power K-Dense-AI ISSUES FOUND Caution 8 1 0 1 0 0
hugging-face-evaluation huggingface SAFE Do not install 100 96 0 0 92 4
hugging-face-paper-publisher huggingface SAFE Caution 50 54 0 0 50 4
hugging-face-jobs huggingface SAFE Do not install 67 8 0 0 5 3
ruzzy trailofbits SAFE Caution 0 8 0 0 3 5
ClawDoro clawhub.ai SAFE Caution 25 7 0 0 2 5
dev-browser sawyerhood SAFE Do not install 100 7 0 0 4 3
vestaboard clawhub.ai SAFE Do not install 53 7 0 0 2 5
agent-task-manager clawhub.ai SAFE Caution 7 6 0 0 3 3
using-git-worktrees skills.sh SAFE Caution 11 6 0 0 2 4
health-tracker clawhub.ai SAFE Caution 2 5 0 0 1 4
subagent-driven-development skills.sh SAFE Caution 0 5 0 0 2 3
redis clawhub.ai SAFE Safe 0 4 0 0 4 0
YT Meta - YouTube Metadata Extractor clawhub.ai SAFE Safe 0 3 0 0 2 1
stealth-browser clawhub.ai SAFE Caution 32 3 0 0 3 0
technews clawhub.ai SAFE Caution 12 3 0 0 3 0
apewisdom clawhub.ai SAFE Caution 7 2 0 0 1 1
aws-agentic-ai zxkane SAFE Do not install 82 2 0 0 1 1
chart-image clawhub.ai SAFE Caution 38 2 0 0 2 0
clawbrowser clawhub.ai SAFE Caution 0 2 0 0 2 0
earnings-calendar clawhub.ai SAFE Do not install 100 2 0 0 1 1
explorer clawhub.ai SAFE Safe 15 2 0 0 1 1
feishu-bridge clawhub.ai SAFE Caution 23 2 0 0 2 0
gemini skills.sh SAFE Do not install 54 2 0 0 2 0
llm clawhub.ai SAFE Safe 0 2 0 0 2 0
openpet clawhub.ai SAFE Safe 0 2 0 0 2 0
ralph-loop clawhub.ai SAFE Caution 16 2 0 0 2 0
spec-to-code-compliance trailofbits SAFE Caution 0 2 0 0 2 0
stock-evaluator-v3 clawhub.ai SAFE Caution 8 2 0 0 2 0
twitter-openclaw clawhub.ai SAFE Caution 49 2 0 0 1 1
azure-mgmt-apicenter-py microsoft SAFE Caution 0 1 0 0 1 0
frontend-ui-dark-ts microsoft SAFE Caution 32 1 0 0 1 0
npkill clawhub.ai SAFE Safe 8 1 0 0 1 0
reddit clawhub.ai SAFE Safe 14 1 0 0 0 1