ClawHub
Back to skill

Security audit

sofya-search

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign verdict with medium confidence: the skill appears purpose-aligned, while its documented use of an external search service raises a privacy question and one referenced path was not resolved.

Before installing, consider whether you are comfortable sending search queries, URLs, and prompts to Sofya. Avoid including confidential information or secrets in inputs. The scan is partial: it could not resolve one local path-like reference, so the evidence is incomplete.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

The workflow sends search queries, URLs, and extraction prompts to Sofya without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The workflow instructs the agent to call Sofya directly using search queries, supplied URLs, and extraction prompts. Although the API destination is documented, there is no warning that these inputs are disclosed to a third-party service and could contain confidential information or access tokens embedded in URLs.

Content

No source excerpt is available for this finding.

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 · SKILL.md (reported line 51)May include surrounding context.

Parameters: query (required), search_depth (snippets | basic), max_results (1-20, default 10), include_answer (bool), topic (general | news), freshness (day, week, month, year, or YYYY-MM-DD:YYYY-MM-DD), include_domains / exclude_domains (bare hostnames like github.com, max 10).

bash
curl -s -X POST https://sofya.co/v1/search \
  -H "Authorization: Bearer $SOFYA_API_KEY" -H "Content-Type: application/json" \
  -d '{"query": "python 3.14 release notes", "max_results": 5}'