Back to skill

Security audit

AIPASS

Security checks for vulnerabilities and agentic risk

Overview

This skill is an instructions-only wrapper for using AIPASS without exposing credentials, and its sensitive behavior is disclosed and purpose-aligned.

Install only if you intend to use AIPASS and trust the separate signed AIPASS app and OpenClaw tool plugin. Before use, verify the AIPASS reference, origin, and requested GitHub resource, and do not paste raw passwords, tokens, cookies, or MFA codes into chat.

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
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

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

Content

Scanner excerpt · SKILL.md (reported line 19)May include surrounding context.

md
## Standalone preview boundary

This Skill contains instructions only. It does not contain executable code, install the AIPASS macOS application or OpenClaw Tool Plugin, access Keychain, or perform a credentialed operation by itself.

Credentialed operations require both a signed AIPASS native application and the compatible AIPASS OpenClaw Tool Plugin. If an `aipass_*` tool named below is unavailable, stop and explain that the AIPASS Tool Plugin and local application must be installed. Never substitute a shell command, generic HTTP tool, browser automation, environment variable, or raw credential.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

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

Content

Scanner excerpt · SKILL.md (reported line 43)May include surrounding context.

md
When a registered login is required and `aipass_status` has already passed, call `aipass_web_login_handoff` with only the full AIPASS reference and the exact canonical HTTPS origin. Never ask for selectors, an ID, a password, a cookie, or an MFA code.

Stop browser automation after the tool returns `secure_runtime_required`. Explain that AIPASS validated the reference and origin but did not access the credential. The current release requires the user to complete the login manually without placing the ID or password in chat. Do not attempt to observe, race, script, screenshot, or reattach during manual login. Treat MFA, passkeys, CAPTCHA, password changes, federated login, and ambiguous forms as manual blockers.

## GitHub releases and Issue/PR reads

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 23)May include surrounding context.

md
Credentialed operations require both a signed AIPASS native application and the compatible AIPASS OpenClaw Tool Plugin. If an `aipass_*` tool named below is unavailable, stop and explain that the AIPASS Tool Plugin and local application must be installed. Never substitute a shell command, generic HTTP tool, browser automation, environment variable, or raw credential.

Before the first AIPASS operation in a session, call `aipass_status`. Continue only when the tool exists and reports that the local broker is ready. If it is unavailable, unreachable, disabled, incompatible, or returns an unknown state, fail closed without asking for a credential.

## Rules

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% 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 27)May include surrounding context.

md
## Rules

1. Ask for or use only an AIPASS reference such as `aipass://local/github-main`; never ask the user to paste the underlying credential.
2. Call only a service-specific AIPASS tool for the requested operation.
3. Never attempt to read, reveal, print, log, export, transform, encode, or inspect a secret value.
4. Never place credentials in tool arguments, shell commands, environment variables, files, URLs, error reports, or chat messages.