Back to skill

Security audit

Google Developer Knowledge

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with high confidence: the completed scans found no AIG issues, and the flagged items were assessed as expected or contextual for its developer knowledge purpose.

The verdict is benign, with high confidence, and the scan reports complete coverage of the five listed components. Before installing, consider whether you are comfortable sending your queries to the configured MCP service or Google API; avoid including credentials, personal data, or confidential content in queries.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

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 · references/api-fallback.md (reported line 25)May include surrounding context.

OAuth with a Quota Project

Use this alternative when an OAuth access token and quota project are available. Do not persist a short-lived access token in an MCP configuration.

bash
export DEVELOPERKNOWLEDGE_OAUTH_ACCESS_TOKEN="ACCESS_TOKEN"

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 · references/api-fallback.md (reported line 25)May include surrounding context.

OAuth with a Quota Project

Use this alternative when an OAuth access token and quota project are available. Do not persist a short-lived access token in an MCP configuration.

bash
export DEVELOPERKNOWLEDGE_OAUTH_ACCESS_TOKEN="ACCESS_TOKEN"

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 · test-prompts.json (reported line 59)May include surrounding context.

json
{
      "id": "oauth-rest-fallback-and-budget",
      "mode": "full_test",
      "prompt": "沒有可用 MCP,但有 OAuth access token 和 quota project;查 Cloud Run minimum instances 計費並附來源。",
      "assertions": [
        "Uses the REST OAuth fallback with an Authorization header and x-goog-user-project header.",
        "Retrieves no more than three parent documents unless broad extraction was explicitly requested.",

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-fallback.md (reported line 31)May include surrounding context.

md
export DEVELOPERKNOWLEDGE_OAUTH_ACCESS_TOKEN="ACCESS_TOKEN"
export GOOGLE_CLOUD_QUOTA_PROJECT="YOUR_PROJECT_ID"

curl --get "https://developerknowledge.googleapis.com/v1/documents:searchDocumentChunks" \
  --data-urlencode "query=Cloud Run minimum instances billing" \
  -H "Authorization: Bearer $DEVELOPERKNOWLEDGE_OAUTH_ACCESS_TOKEN" \
  -H "x-goog-user-project: $GOOGLE_CLOUD_QUOTA_PROJECT"

The workflow sends query text to the configured MCP service without warning users about disclosure of potentially sensitive query content.

Medium
Category
Output Handling
Confidence
86% confidence
Finding

The search helper transmits its query argument through the configured MCP route; the conceptual-query helper at lines 98–102 does the same. Although the file advises using trusted servers and protecting gateway secrets, it does not warn that queries—including pasted code or error messages—are disclosed to the gateway and upstream service.

Content

No source excerpt is available for this finding.