Back to skill

Security audit

RSS监控

Security checks for vulnerabilities and agentic risk

Overview

This RSS monitoring skill is coherent and purpose-aligned, but users should treat feed URLs and the Python dependency as normal network and supply-chain trust decisions.

Install the dependency in an isolated environment when practical, prefer a pinned feedparser version, and only add public RSS/Atom feed URLs you trust. Do not expose this sample directly as a public URL-fetching service without validation and network egress restrictions.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (2)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:3
Finding

Unpinned Third-Party Dependency Installation

Content
View full analysis
Remediation
View remediation
`. 2. Maintain a lockfile containing resolved transitive dependencies. 3. Require package hashes, such as with a hashed requirements file and `pip install --require-hashes`. 4. Explicitly use a trusted package index and prevent unintended fallback to untrusted indexes. 5. Install the dependency in an isolated virtual environment under a non-privileged account. 6. Add automated dependency vulnerability and integrity scanning. 7. Review and deliberately update the pinned version rather than automatically consuming the latest release. ]]>

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:33
Finding

Unrestricted Feed URL Fetching Enables Server-Side Request Forgery

Content
View full analysis
list: updates = [] for name, url in self.feeds.items(): feed = feedparser.parse(url) for entry in feed.entries[:5]: entry_time = datetime(*entry.published_parsed[:6]) if name not in self.last_check or entry_time > self.last_check[name]: updates.append({ 'feed': name, 'title': entry.title, 'link': entry.link, 'published': entry.get('published', 'Unknown'), }) self.last_check[name] = datetime.now() return updates def get_entries(self, url: str, limit: int = 10): feed = feedparser.parse(url) return [{ 'title': e.title, 'link': e.link, 'summary': e.get('summary', '')[:200], } for e in feed.entries[:limit]] ``` ### Technical Analysis The `add_feed` and `get_entries` methods accept caller-controlled locations and pass them directly to `feedparser.parse`. No validation restricts the URI scheme, hostname, resolved IP address, port, redirects, or destination network. When the caller can influence the URL, the runtime may be induced to request resources from destinations accessible to the agent host but inaccessible to the caller. Potential targets include loopback services, private-network services, link-local cloud metadata endpoints, and other internal HTTP resources. Depending on the schemes supported by the installed parser and runtime, non-network resource locations may also require explicit rejection. The methods return parsed titles, links, and summaries. If an internal endpoint returns content that can be interpreted as a f ...[truncated 1915 chars]
Remediation
View remediation
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The natural-language description and activation instructions are presented in Chinese only, with no indication that other languages are supported or that the user can opt in to this locale. This can violate language/locale policy because it implicitly constrains usage to a specific language without documented justification or user choice.

Content

No source excerpt is available for this finding.