Back to skill

Security audit

Tetrac Perp Trader

Security checks for vulnerabilities and agentic risk

Overview

This skill is a real trading assistant, but it routes reusable exchange credentials through a central proxy and exposes high-impact account actions that need careful review before installation.

Install only if you trust TTC Box with exchange API credentials and understand that connected keys may allow trading, position closure, leverage changes, and possibly withdrawals if over-permissioned. Use trade-only exchange API keys with withdrawals disabled, avoid the credential-bearing GET pattern, do not print .env secrets during troubleshooting, prefer the OpenClaw installer over unpinned npx, and treat this as a high-impact financial integration rather than a simple market-data tool.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (5)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
references/api-reference.md:84
Finding

Reusable Exchange Credentials Are Transmitted to a Central Proxy

Content
View full analysis
", "method": "", "params": {}, "credentials": { "apiKey": "...", "apiSecret": "...", "passphrase": "...", "walletAddress": "..." } } ``` ``` The architecture is explicitly described in `references/exchanges.md:3`: ```markdown All exchanges are accessed via the TTC Box proxy at `https://ttc.box`. You never call exchange APIs directly. ``` ### Technical Analysis Every authenticated exchange operation is routed through TTC Box and includes reusable exchange credentials in the request body. This behavior is disclosed rather than covert, but it establishes TTC Box as a centralized trust boundary with access to credentials capable of reading account data and performing trading operations. Sending complete exchange API secrets to an intermediary is not the minimum privilege required for a trading client. A less privileged design would sign exchange requests locally or use delegated, narrowly scoped credentials that cannot be replayed by the proxy. The supplied artifact contains only documentation, so the proxy's credential retention, logging, encryption, isolation, and deletion controls cannot be verified. ### Attack Path 1. A user stores exchange API keys, secrets, and passphrases in the local environment. 2. The client sends those credentials to `https://ttc.box` when an exchange operation is requested. 3. An attacker compromises TTC Box, its request logs, application monitoring, administrative accounts, or credential ...[truncated 748 chars]
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Error
Location
references/api-reference.md:108
Finding

Proxy API Exposes Withdrawal Capability Beyond the Skill's Declared Trading Scope

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
references/api-reference.md:49
Finding

Plain Passkey Transmission and Query-String API Secret Exposure

Content
View full analysis
" } ``` ``` The same API documentation also defines a GET request containing exchange credentials: ```markdown #### Get trade history (GET variant) ``` ```text GET /api/v1/exchanges?method=getUserTradeHistory&exchange=bybit&apiKey=...&apiSecret=... ``` ### Technical Analysis The login process sends the user's passkey to the server in recoverable form at the application layer. TLS protects it while in transit but does not prevent exposure at the destination, in request-body logging, tracing systems, crash reports, or a compromised server. Hashing only after receipt does not provide client-side secrecy. The GET API variant is additionally unsafe because it embeds `apiKey` and `apiSecret` in a URL. URLs are routinely retained in reverse-proxy logs, web-server access logs, browser history, observability platforms, caches, copied diagnostics, and referrer data. Credentials should never be passed in query strings. ### Attack Path 1. The user logs in, causing the plain passkey to be sent to TTC Box. 2. Alternatively, an integration uses the documented GET variant and places exchange credentials in the URL. 3. A reverse proxy, application server, monitoring tool, support bundle, browser, or compromised endpoint records the request. 4. An attacker with access to those records extracts the passkey or exchange secret. 5. The attacker uses the passkey to attempt TTC authentication or uses the exchange credentials to access the user's trading account. 6. The attacker performs operations allowed by the compromised cred ...[truncated 395 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
references/troubleshooting.md:154
Finding

Troubleshooting Command Prints the Full Passkey to Agent and Terminal Logs

Content
View full analysis
` and `TTC_PASSKEY=`. 4. The execution framework records or forwards terminal output. 5. An attacker, support operator, telemetry consumer, or unauthorized log reader obtains the passkey. 6. The attacker attempts authentication or other wallet-related operations using the exposed secret. ### Impact Assessment The issue discloses personally identifying account data and a high-value authentication or wallet-encryption secret. The effective exposure scope includes everyone with access to agent transcripts, terminal logs, o ...[truncated 100 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
README.md:46
Finding

Installation Command Executes an Unpinned Mutable Package Release

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The GET example places sensitive exchange API credentials directly in the URL query string, which can be exposed through browser history, terminal history, reverse proxies, access logs, analytics systems, and referrer leakage. Because this skill operates a multi-exchange trading tool handling real account credentials, leaked API keys and secrets could enable unauthorized trading, data access, or withdrawals depending on exchange permissions.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/troubleshooting.md (reported line 40)May include surrounding context.

Proactive check: Before starting any session, check token age:

bash
issued=$(grep TTC_TOKEN_ISSUED_AT .env | cut -d= -f2)
age=$(( $(date +%s) - issued ))
echo "Token age: ${age}s / 86400s"

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/troubleshooting.md (reported line 161)May include surrounding context.

Proactive check: Before starting any session, check token age:

bash
issued=$(grep TTC_TOKEN_ISSUED_AT .env | cut -d= -f2)
age=$(( $(date +%s) - issued ))
echo "Token age: ${age}s / 86400s"

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/troubleshooting.md (reported line 165)May include surrounding context.

Proactive check: Before starting any session, check token age:

bash
issued=$(grep TTC_TOKEN_ISSUED_AT .env | cut -d= -f2)
age=$(( $(date +%s) - issued ))
echo "Token age: ${age}s / 86400s"

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The README instructs installation via npx clawhub@latest install tetrac-perp-trader, which executes remote package code without pinning to a specific version. If the upstream package, dependency chain, or publishing account is compromised, users or agents may run attacker-controlled code during install. In this skill’s context, that risk is more serious because the tool handles trading operations and environment credentials, so install-time compromise could expose tokens or tamper with binaries.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 678)May include surrounding context.

md
### Read vs write operations

Read operations are idempotent — re-running them is safe. Write operations mutate exchange state and CANNOT be safely re-run without verification.

| Read (idempotent — retry freely) | Write (NOT idempotent — verify before any retry) |
|---|---|

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documentation advertises destructive financial operations such as withdrawals, closing all positions, and canceling all orders without prominent warnings, confirmations, or guidance about irreversible trading consequences. In a trading CLI context, this increases the likelihood of accidental misuse by users or downstream agents, potentially causing unintended losses or fund movements.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This markdown file includes TTC auth tokens, API keys, API secrets, passphrases, and wallet-related values in a .env example, but it does not warn users that these are sensitive credentials that must be protected and not committed or shared. Because the file is markdown, SQP-2 applies when the description omits warnings about behaviors affecting user data or privacy.

Content

No source excerpt is available for this finding.