Back to skill

Security audit

VCF Log Insight Health Check

Security checks for vulnerabilities and agentic risk

Overview

The skill does the expected Log Insight health checks, but it sends an API bearer token while disabling TLS certificate verification.

Install only if you trust the configured LOGINSIGHT_HOST and understand that the current commands disable TLS certificate validation. Prefer removing -k, using a trusted CA bundle for private certificates, and using a read-only token limited to health and node-status endpoints.

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 (1)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:22
Finding

TLS Certificate Verification Disabled for Authenticated API Requests

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 22-31
Vulnerability Type: TLS certificate validation bypass resulting in potential bearer-token disclosure
Risk Level: High

Vulnerable Code

bash
curl -s -k -X GET "https://$LOGINSIGHT_HOST/api/v1/health" \
  -H "Authorization: Bearer $LOGINSIGHT_API_TOKEN" \
  -H "Accept: application/json" | jq '.'
bash
curl -s -k -X GET "https://$LOGINSIGHT_HOST/api/v1/cluster/nodes" \
  -H "Authorization: Bearer $LOGINSIGHT_API_TOKEN" \
  -H "Accept: application/json" | jq '.nodes[] | {id, ip, status, uptime}'

Technical Analysis

Both authenticated requests use the curl -k option, which disables TLS certificate-chain and hostname verification. The bearer token is therefore sent over an encrypted connection without reliable authentication of the remote server.

Network access and bearer-token authentication are necessary for the Skill's declared health-check functionality. There is no evidence of intentional credential exfiltration. However, disabling certificate validation is not necessary and violates minimum-privilege and secure-transport principles. An attacker capable of intercepting or redirecting traffic could present an arbitrary certificate, impersonate the configured Log Insight server, and receive the Authorization header.

Attack Path

  1. The Agent reads LOGINSIGHT_HOST and LOGINSIGHT_API_TOKEN from its environment.
  2. An attacker gains a network interception or redirection position, such as through DNS poisoning, routing manipulation, or a hostile proxy.
  3. The attacker presents a certificate that is untrusted or does not match LOGINSIGHT_HOST.
  4. Because curl -k suppresses certificate and hostname validation errors, the Agent establishes the connection.
  5. The Agent sends Authorization: Bearer $LOGINSIGHT_API_TOKEN to the impersonated endpoint.
  6. The attacker captures the token and can also return fabric ...[truncated 673 chars]
Remediation
View remediation

Remediation Suggestions

  1. Remove -k from both curl commands so certificate-chain and hostname validation remain enabled.
  2. For environments using a private certificate authority, supply a securely provisioned CA bundle with --cacert /trusted/path/ca.pem rather than disabling validation globally.
  3. Restrict LOGINSIGHT_HOST to an administrator-approved hostname and reject unexpected URL components or untrusted runtime overrides before transmitting credentials.
  4. Use a dedicated, read-only API token limited to health and node-status endpoints.
  5. Rotate the token if these commands have been used across networks where interception may have been possible.
  6. Avoid verbose logging or shell tracing that could expose the Authorization header.
  7. Fail closed when TLS validation fails and report the certificate error rather than retrying insecurely.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 13)May include surrounding context.

md
- LOGINSIGHT_HOST
        - LOGINSIGHT_API_TOKEN
      bins:
        - curl
        - jq
    primaryEnv: LOGINSIGHT_API_TOKEN
    emoji: ""

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
93% confidence
Finding

The command uses curl with -k, which disables TLS certificate validation while sending a bearer token in the Authorization header. That makes the request susceptible to man-in-the-middle interception or redirection, and because LOGINSIGHT_HOST comes from the environment, a misconfigured or attacker-controlled host value could cause the token to be transmitted to an untrusted endpoint.

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

text
3. **Fetch Node Status:** Use the `exec` tool to run the following command to check individual node uptime and status:
   ```bash
   curl -s -k -X GET "https://$LOGINSIGHT_HOST/api/v1/cluster/nodes" \
     -H "Authorization: Bearer $LOGINSIGHT_API_TOKEN" \
     -H "Accept: application/json" | jq '.nodes[] | {id, ip, status, uptime}'

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill instructs the agent to read a bearer token from the environment and send it over the network, but it does not disclose this sensitive action to the user or require explicit confirmation. This creates a secret-handling and transparency risk: users may trigger the skill without realizing credentials are being used, and any compromise of the target host or request path would expose a valid API token.

Content

No source excerpt is available for this finding.