Back to skill

Security audit

Versatile Map Maker

Security checks for vulnerabilities and agentic risk

Overview

This map-making skill is mostly purpose-aligned, but it should be reviewed because it can produce unsafe SVG output from unvalidated style or input files and recommends unpinned package-manager execution.

Review before installing or using with untrusted files. Prefer pinned dependencies or a locked environment, avoid running unpinned npx commands, and only process style JSON, GeoJSON, and SVG inputs from sources you trust until SVG color validation and sanitization are added.

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)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/recolor_choropleth.py:34
Finding

Unvalidated style values allow active-content injection into generated SVG files

Content
View full analysis
' stroke = "#7a7a7a" if rid in region_ids_colors else style["neutral_stroke"] tag = re.sub(r'stroke="[^"]*"', f'stroke="{stroke}"', tag) if 'stroke="' in tag else tag[:-1] + f' stroke="{stroke}">' return tag ``` ```python font = html.escape(style["font_family"]) return f''' {stops} {html.escape(label)} ``` ```python rows.append(f'{html.escape(str(cat))}') ``` ```python extras = f''' {html.escape(args.title)} {html.escape(args.subtitle)} {legend_svg} ''' ``` ### Technical Analysis The `--style` option loads arbitrary JSON and merges every supplied property int ...[truncated 2154 chars]
Remediation
View remediation
` elements - `on*` event-handler attributes - `` elements - Unsafe `href` or `xlink:href` values - Unapproved external-resource references 8. Add regression tests using values containing quotes, angle brackets, event handlers, URLs, and malformed CSS. 9. Document that untrusted SVG and style files must not be processed until sanitization is implemented. ]]>

T08 · Insecure Dependencies

Warning
Location
references/geometry-inputs.md:13
Finding

Unpinned third-party package retrieval creates a mutable supply-chain execution path

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill description presents a map-making utility, but the body includes additional behavior to fetch external river data, cache downloads locally, and produce intermediate JSON geometry artifacts. This mismatch can bypass policy review or user expectations, because operators may approve it for SVG generation without realizing it performs broader network retrieval and data processing actions.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding

The skill instructs use of network access, shell commands, package installation, and filesystem reads/writes but does not declare any explicit tool scope or permissions boundaries. That makes the operational privilege of the skill ambiguous and increases the chance it will be run with broader capabilities than necessary, enabling unnecessary external access or local file modification if invoked in a permissive environment.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

The documentation recommends running npx topojson-client without pinning a specific package version. npx may fetch the latest package from the registry at execution time, which creates a supply-chain risk: a compromised upstream release, typo/confusion, or unexpected breaking change could result in unreviewed code being executed on the user's machine. In this skill context, the command is presented as an operational conversion step, making it more likely a user would copy-paste and run it directly.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/fetch_base_map.py (reported line 43)May include surrounding context.

python
if not shutil_which("npm"):
        raise RuntimeError("npm is required for the Highcharts fast path. Use a supplied GeoJSON instead.")
    print(f"Downloading {PKG} (first run only, cached after this)...", file=sys.stderr)
    subprocess.run(
        ["npm", "pack", PKG, "--silent", "--pack-destination", str(cache_dir)],
        check=True, cwd=cache_dir,
    )

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

This code performs a file write to the path supplied by the user, but there is no confirmation prompt and no explicit warning comment or docstring disclosing that the script will overwrite or create the output SVG. For code files, file writes can be flagged when they lack any form of user disclosure, and the top-level docstring does not mention writing output files.

Content

No source excerpt is available for this finding.