Back to skill

Security audit

Brouter Stake

Security checks for vulnerabilities and agentic risk

Overview

This skill is openly about staking real BSV, but it includes broad live money-moving and paid-oracle API instructions without enough guardrails.

Review before installing. Only use this with a Brouter account and token you are willing to let an agent use for real BSV staking or related API actions. Confirm the market, outcome, amount, and loss risk before any POST that stakes, creates markets, votes, publishes signals, claims faucet funds, or sends payment headers. Keep bearer tokens out of logs and treat the X-Payment and commit-reveal examples as documentation that needs stronger production safeguards.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
references/api.md:270
Finding

Paid Oracle Signals Are Released Before On-Chain Payment Verification

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
references/api.md:369
Finding

Predictable Commit-Reveal Salt Allows Offline Vote Recovery

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (12)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly facilitates real-money staking of BSV sats, including examples that place bets, but it does not warn users that funds are deducted immediately and may be lost. In a transactional betting context, omission of a clear loss-risk warning can mislead users into taking irreversible financial actions without informed consent.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This command performs an authenticated POST to an external API that places a real-money stake, causing immediate transfer of funds based on user-provided parameters. Because it is a live financial action, accidental, spoofed, or insufficiently confirmed execution could result in irreversible loss of funds.

Content

Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.

md
curl -s "$BASE/api/markets?state=OPEN" | jq '.data.markets[] | {id, title, tier}'

# Take a position (minimum 100 sats)
curl -sX POST $BASE/api/markets/{market-id}/stake \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"outcome":"yes","amountSats":100}' | jq .

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This is another authenticated external POST example that directly executes a real-money staking transaction. In the context of a betting skill, the danger is elevated because the action is financially irreversible and could be triggered by misunderstanding, prompt injection into an agent workflow, or lack of user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 78)May include surrounding context.

Stake

bash
curl -sX POST $BASE/api/markets/{market-id}/stake \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"outcome":"yes","amountSats":500}'

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The onboarding example immediately performs registration, obtains a bearer token, and executes a real-money staking action without warning that the token is sensitive and the action can affect balances. In this skill context—real BSV staking—that omission is especially risky because an agent may replay the workflow automatically and expose credentials or spend funds without informed user approval.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/api.md (reported line 9)May include surrounding context.

md
BASE=https://brouter.ai

# 1. Register (get your token + 5000 sats)
curl -sX POST $BASE/api/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name":"youragent","publicKey":"02your33bytepubkeyhex"}' | jq .

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/api.md (reported line 480)May include surrounding context.

md
BASE=https://brouter.ai

# 1. Register (get your token + 5000 sats)
curl -sX POST $BASE/api/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name":"youragent","publicKey":"02your33bytepubkeyhex"}' | jq .

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/api.md (reported line 17)May include surrounding context.

md
curl -s "$BASE/api/markets?state=OPEN" | jq '.data.markets[0].id'

# 3. Stake on a market (use token from step 1)
curl -sX POST $BASE/api/markets/{market-id}/stake \
  -H "Authorization: Bearer {your-token}" \
  -H "Content-Type: application/json" \
  -d '{"outcome":"yes","amountSats":100}'

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The reference file documents capabilities to create markets and interact with oracle-driven resolution that go beyond the skill’s declared purpose of browsing and staking on markets. This scope expansion is dangerous because an integrating agent may expose higher-risk actions—such as creating binding market objects or invoking resolution-related workflows—without the user expecting those powers from the manifest.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This section introduces monetized oracle publication, paid signal sales, and custom payment-header construction, which materially exceed a simple staking/browsing use case. The mismatch is dangerous because it can lead an agent to handle monetization flows, payment proofs, and token-authenticated publishing operations that were not disclosed in the skill metadata, increasing the chance of unintended fund movement, secret misuse, or unsafe automation.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/api.md (reported line 486)May include surrounding context.

md
TOKEN=$(echo $RESP | jq -r '.data.token')

# 2. Claim faucet
curl -sX POST $BASE/api/agents/alice/faucet -H "Authorization: Bearer $TOKEN"

# 3. Create market
MID=$(curl -sX POST $BASE/api/markets \

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/api.md (reported line 496)May include surrounding context.

md
| jq -r '.data.market.id')

# 4. Stake
curl -sX POST $BASE/api/markets/$MID/stake \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"outcome":"yes","amountSats":100}'

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill instructs users to send bearer-token-authenticated requests to an external service but does not warn that account-linked identifiers, positions, and transaction metadata will be transmitted to brouter.ai. This creates a privacy and account-security concern, especially if users do not understand that authenticated actions can reveal identity, balances, or trading history.

Content

No source excerpt is available for this finding.