Back to skill

Security audit

行业景气全景-cn

Security checks for vulnerabilities and agentic risk

Overview

This reporting skill needs review because its connection checks can expose your API key beyond the destinations its safeguards promise.

Use user-supplied materials without a Cue key until credential handling is corrected. The shell example should restrict destinations to trusted HTTPS hosts, and the Python helper should reject redirects or validate every hop before forwarding authentication. A truly offline diagnostic requires --catalog-file without --post. Review immutable installer/source versions and optional integrations separately; their runtime behavior and the external service were not tested.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (23)

Tainted flow: 'req' from os.environ.get (line 188, 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/cue_connectivity_check.py (reported line 193)May include surrounding context.

python
code = None
    raw = ""
    try:
        with urllib.request.urlopen(req, timeout=args.timeout) as resp:
            code, raw = resp.status, resp.read().decode("utf-8", "replace")
    except urllib.error.HTTPError as err:
        detail = ""

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · README.en.md (reported line 47)May include surrounding context.

bash
CATALOG="https://cuecue.cn/api/mcp-catalog"
curl -s "$CATALOG" | python3 -c 'import json,sys
rows=json.load(sys.stdin)["data"]["domains"]
[print(r.get("id"), r.get("external_status"), (r.get("routing") or {}).get("url", "-")) for r in rows]'
DOMAIN_ID="<replace with one of the ids listed by the line above>"

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · README.en.md (reported line 51)May include surrounding context.

md
rows=json.load(sys.stdin)["data"]["domains"]
[print(r.get("id"), r.get("external_status"), (r.get("routing") or {}).get("url", "-")) for r in rows]'
DOMAIN_ID="<replace with one of the ids listed by the line above>"
ROUTING_URL="$(curl -s "$CATALOG" | python3 -c 'import json,sys
rows=json.load(sys.stdin)["data"]["domains"]
hit=[r["routing"]["url"] for r in rows if r.get("id")==sys.argv[1] and r.get("routing")]
print(*hit,sep="\n") if hit else sys.exit("Not found: that id is not in the catalog response (or its routing has no url field) - run the line above and pick one of the ids listed")' "$DOMAIN_ID")"

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · README.md (reported line 51)May include surrounding context.

bash
CATALOG="https://cuecue.cn/api/mcp-catalog"
curl -s "$CATALOG" | python3 -c 'import json,sys
rows=json.load(sys.stdin)["data"]["domains"]
[print(r.get("id"), r.get("external_status"), (r.get("routing") or {}).get("url", "-")) for r in rows]'
DOMAIN_ID="<把这里换成上一行列出的那个域 id>"

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · README.md (reported line 55)May include surrounding context.

md
rows=json.load(sys.stdin)["data"]["domains"]
[print(r.get("id"), r.get("external_status"), (r.get("routing") or {}).get("url", "-")) for r in rows]'
DOMAIN_ID="<把这里换成上一行列出的那个域 id>"
ROUTING_URL="$(curl -s "$CATALOG" | python3 -c 'import json,sys
rows=json.load(sys.stdin)["data"]["domains"]
hit=[r["routing"]["url"] for r in rows if r.get("id")==sys.argv[1] and r.get("routing")]
print(*hit,sep="\n") if hit else sys.exit("未取到:该 id 不在目录现值里(或它没有 routing 里的 url 字段)——先跑上面那行列出现场的 id 再填")' "$DOMAIN_ID")"

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The supplied code's implemented purpose is service discovery and authenticated connectivity verification, not the declared industry-report workflow. It contains no industry/time-window analysis, evidence-series processing, company selection, or report assembly. Connectivity checks could support a larger reporting skill, but the supplied description does not establish that relationship or disclose this standalone credential-bearing diagnostic capability. The POST is opt-in and its initial destination is checked for HTTPS and an authorized hostname suffix; these safeguards reduce risk but do not resolve the purpose mismatch. Without --post, the script still performs an anonymous catalog GET unless a local snapshot is supplied.

Content

No source excerpt is available for this finding.

Automatic redirects can bypass the promised credential-destination checks.

High
Category
Excessive Agency
Confidence
99% confidence
Finding

The guard validates only the initial URL, while urllib's default redirect handler can follow redirects and preserve the explicitly supplied Authorization header, including on a POST-to-GET redirect. An authorized endpoint that is compromised or exposes an applicable redirect can therefore send the bearer token to an unauthorized host or an HTTP destination, bypassing both promised checks.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The active quick-start command invokes an unversioned npm installer and installs skill content from an unpinned repository revision. Although this is not an MCP-server declaration, it creates a supply-chain exposure: a compromised future installer or repository update could execute code or introduce hostile agent instructions.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The individual evidence-leg installation command selects a skill but still does not pin the installer or source revision. A later compromised release could therefore introduce executable installer code or malicious skill instructions into an agent environment that may subsequently access credentials.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The whole-collection installation command leaves both the npm installer version and the collection revision unspecified. Installing a broader collection increases the amount of mutable third-party skill content entering the agent's environment; describing it as official does not provide integrity protection.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
99% confidence
Finding

The shell example attaches CUE_API_KEY to a URL taken directly from the catalog without checking its scheme or destination host. A compromised or malicious catalog entry could direct the credential to an attacker-controlled endpoint or transmit it over HTTP; the protections described for the separate Python script do not protect this curl command.

Content

Scanner excerpt · README.en.md (reported line 56)May include surrounding context.

md
hit=[r["routing"]["url"] for r in rows if r.get("id")==sys.argv[1] and r.get("routing")]
print(*hit,sep="\n") if hit else sys.exit("Not found: that id is not in the catalog response (or its routing has no url field) - run the line above and pick one of the ids listed")' "$DOMAIN_ID")"
[ -n "$ROUTING_URL" ] || { echo "Fetch failed: ROUTING_URL is empty - run the line above, then set DOMAIN_ID to one of the ids listed" >&2; exit 1; }
curl -s -X POST "$ROUTING_URL" \
  -H "Authorization: Bearer $CUE_API_KEY" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' \

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The installation command leaves both the npx package and the downloaded skill revision unpinned, allowing future upstream changes to alter what is executed or installed. This is a supply-chain exposure rather than an unversioned MCP server specifically; checking that the skill appears after installation does not verify its integrity.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The collection installation command resolves an unpinned npx package and mutable upstream skill content. Installing supporting data-access skills makes upstream compromise consequential because those skills may later operate with service credentials.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The single-skill installation example also lacks immutable versions for the installer and repository content. Selecting one named skill narrows installation scope but does not prevent a changed or compromised upstream revision from supplying different instructions or code.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
99% confidence
Finding

The shell example sends CUE_API_KEY to a URL taken directly from a remote catalog without validating its scheme or destination host. A compromised catalog could direct the bearer token to an attacker or an unencrypted HTTP endpoint; the separate Python helper's destination checks do not protect this copyable shell path.

Content

Scanner excerpt · README.md (reported line 60)May include surrounding context.

md
hit=[r["routing"]["url"] for r in rows if r.get("id")==sys.argv[1] and r.get("routing")]
print(*hit,sep="\n") if hit else sys.exit("未取到:该 id 不在目录现值里(或它没有 routing 里的 url 字段)——先跑上面那行列出现场的 id 再填")' "$DOMAIN_ID")"
[ -n "$ROUTING_URL" ] || { echo "取值失败:ROUTING_URL 为空——先跑上面那行列出现场 id,再把 DOMAIN_ID 换成其中之一" >&2; exit 1; }
curl -s -X POST "$ROUTING_URL" \
  -H "Authorization: Bearer $CUE_API_KEY" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' \

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

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.

Manifest references MCP server without version pin: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The manifest includes the same unpinned installation recommendation, leaving installation dependent on mutable upstream content. Its own version field identifies this skill release but does not pin what the recommended installation command downloads.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The description recommends an installation command that resolves mutable installer and repository versions. Although this is not an MCP server declaration, following the command exposes the installation to unreviewed upstream changes.

Content

No source excerpt is available for this finding.

The claimed zero-network default still fetches the remote catalog.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

Without --catalog-file, the program fetches the remote catalog before checking whether --post was supplied, contradicting its repeated zero-network-default claims. Users relying on those claims can unintentionally disclose their network origin and contact an external service, although this default GET does not attach the API key.

Content

No source excerpt is available for this finding.