Security audit

Indices

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Indices CLI reference skill with credential and secret-management risks that are disclosed enough to review carefully but not evidence of hidden or malicious behavior.

Install only if you intend to let the agent help with Indices account operations. Prefer interactive login and secret prompts, avoid putting real API keys or secrets directly in command arguments, review any delete or --yes operation before it runs, and do not allow task creation payloads that enable fully autonomous execution.

A.I.G

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

Warning

Location: SKILL.md:23
Finding: API Key Exposed Through Command-Line Arguments
Content: View full analysis
" # non-interactive ``` ### Technical Analysis The documented non-interactive login method places an API key directly in the command-line argument list. Depending on the operating system, shell, and execution environment, command-line arguments may be exposed through: - Shell history files - Process inspection facilities - Terminal session recordings - Agent or automation execution logs - Monitoring and auditing systems - Error reports that capture invoked commands Although the document also provides a secure interactive login command, the unsafe argument-based method remains explicitly documented and may be selected for automation. ### Attack Path 1. A user or Agent follows the documented non-interactive login example. 2. The API key is expanded into the command line as plaintext. 3. The shell, Agent framework, process monitor, or logging system records or exposes the command. 4. A local user or log reader retrieves the API key. 5. The exposed credential is used to authenticate to the Indices service, subject to the permissions and validity of that API key. ### Impact Assessment Successful exploitation discloses the user's Indices API credential. An attacker may obtain the remote privileges assigned to that key, potentially including access to authentication-protected task, run, and secret-management operations. The precise scope is limited by the service-side permissions, expiration, and revocation state of the compromised credential. ]]>
Remediation: View remediation
"` example from the recommended workflow. - Prefer `indices login`, which obtains the key through a secure interactive prompt. - For non-interactive automation, support a protected input channel such as an inherited file descriptor or standard input without echoing or logging its contents. - If environment-variable support is introduced, ensure automation systems mask the variable and do not print the environment in logs. - Configure the CLI and Agent runtime to redact credential-bearing arguments from command traces, diagnostics, and audit logs. - Advise users to rotate any API key that may already have appeared in shell history or execution logs. - Use narrowly scoped, short-lived API credentials where the service supports them. ]]>

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:96
Finding: Secret Value Exposed Through Command-Line Arguments
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The activation text says to use this skill for 'any operation with the Indices CLI,' which is very broad and may cause an agent to invoke a powerful skill in situations that involve authentication, secret handling, deletion, or task/run operations without sufficient narrowing. Overly broad routing increases the chance of unnecessary exposure to sensitive operations and accidental misuse rather than creating a direct exploit by itself.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The skill explicitly says never to set is_fully_autonomous to true, but it still provides unrestricted generic task-creation pathways via --creation-params, --body, --file, and stdin that could be used to supply prohibited fields anyway. Because the skill is documentation rather than an enforcing wrapper, an agent following it could still create autonomous tasks if it passes raw JSON, undermining the stated safety boundary.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.