Back to skill

Security audit

小红书笔记批量导出 Xiaohongshu batch export

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: the reported behavior is broadly consistent with the skill’s export purpose, though its declared purpose and some capabilities do not fully match, and scan coverage is partial.

The verdict is benign, not a guarantee of zero risk. Consider whether the description and reported capabilities fit your needs, and whether you are comfortable with network use and writing exported files. Confidence is medium, and coverage is partial because two referenced local paths could not be matched to bundled artifacts; treat the evidence as incomplete.

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
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (4)

Tainted flow: 'req' from os.environ.get (line 103, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/linkdigest.py (reported line 111)May include surrounding context.

python
if data is not None:
        req.add_header("Content-Type", "application/json")
    try:
        with urllib.request.urlopen(req, timeout=timeout, context=_SSL["context"]) as resp:
            status, headers, raw = resp.status, resp.headers, resp.read()
    except urllib.error.HTTPError as err:
        status, headers, raw = err.code, err.headers, err.read()

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The client supports the declared service integration, but the supplied chunk alone does not implement the central batch-export workflow and includes additional user-facing operations. It could serve as a helper within a larger matching skill, but that orchestration is not shown. Reading LINKDIGEST_API_KEY and sending supplied text to LinkDigest are expected supporting behaviors, not suspicious credential access. No Xiaohongshu cookie access, media downloads, local export writes, or unrelated data collection are evident.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, network, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Tainted flow: 'name' from os.environ.get (line 113, credential/environment) → open (file write)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/batch_export.py (reported line 114)May include surrounding context.

python
continue
        title = d.get("title") or (d.get("caption") or "").split("\n")[0][:40] or "(无标题)"
        name = f"{n:02d}-{safe_name(title)}.md"
        with open(os.path.join(args.out, name), "w", encoding="utf-8") as f:
            f.write(d.get("raw_markdown") or json.dumps(d, ensure_ascii=False, indent=2))
        likes = stat(d, "likes", "like")
        rows.append(f"| {n} | {title} | {d.get('author') or ''} | {d.get('posted_at') or ''} | {likes} | [{name}]({name}) |")