Back to skill

Security audit

air-chart

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this air-chart skill suspicious with medium confidence, chiefly because its optional installation instructions use an unpinned installer and parts of two scripts were only partially inspected.

The verdict is suspicious, not malicious, and the scan is incomplete. Before installing, consider whether you trust the source and are comfortable with an unpinned install command and the possibility of misleading chart comparisons; the available scan does not confirm malicious behavior or establish that every component is safe.

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
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (19)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 27)May include surrounding context.

md
scripts/air_chart.sh --city <city> [--city <city2> ...] [--param <pollutant>] [--period <range>] [options]

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 89)May include surrounding context.

md
- `scripts/openaq_client.sh`: API v3 wrapper (X-API-Key auth, User-Agent, md5 cache, 429 exponential backoff, liveness filter, sensor match by fixed parameter I

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.es.md (reported line 59)May include surrounding context.

md
207d%20%C2%B7%20multiciudad%20%C2%B7%20mismo%20par%C3%A1metro&unit=%C2%B5g%2Fm%C2%B3&labels=11-07,11-08,11-09,11-10,11-11,11-12,11-13&series=Tokyo%20PM2.5:2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=OMS%2024h:15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=Tendencia+de+New+Delhi+PM2.5+%C2%B7+%C3%BAltimos+30d+%C2%B7+hora+local+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=OMS+24h%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## Características

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.fr.md (reported line 59)May include surrounding context.

md
s%207d%20%C2%B7%20multi-villes%20%C2%B7%20m%C3%AAme%20param%C3%A8tre&unit=%C2%B5g%2Fm%C2%B3&labels=11-07,11-08,11-09,11-10,11-11,11-12,11-13&series=Tokyo%20PM2.5:2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=OMS%2024h:15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=Tendance+New+Delhi+PM2.5+%C2%B7+derniers+30d+%C2%B7+heure+locale+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=OMS+24h%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## Caractéristiques

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.hi.md (reported line 59)May include surrounding context.

md
s=Tokyo%20PM2.5:2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=WHO%2024h:15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=New+Delhi+PM2.5+%E0%A4%B0%E0%A5%81%E0%A4%9D%E0%A4%BE%E0%A4%A8+%C2%B7+%E0%A4%AA%E0%A4%BF%E0%A4%9B%E0%A4%B2%E0%A5%87+30d+%C2%B7+%E0%A4%B8%E0%A5%8D%E0%A4%A5%E0%A4%BE%E0%A4%A8%E0%A5%80%E0%A4%AF+%E0%A4%B8%E0%A4%AE%E0%A4%AF+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=WHO+24h%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## विशेषताएं

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.ja.md (reported line 59)May include surrounding context.

md
it=%C2%B5g%2Fm%C2%B3&labels=11-07,11-08,11-09,11-10,11-11,11-12,11-13&series=Tokyo%20PM2.5:2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=WHO%2024%E6%99%82%E9%96%93:15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=New+Delhi+PM2.5+%E3%81%AE%E6%8E%A8%E7%A7%BB+%C2%B7+%E9%81%8E%E5%8E%BB+30d+%C2%B7+%E7%8F%BE%E5%9C%B0%E6%99%82%E9%96%93+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=WHO+24%E6%99%82%E9%96%93%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## 特徴

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.ko.md (reported line 59)May include surrounding context.

md
%98&unit=%C2%B5g%2Fm%C2%B3&labels=11-07,11-08,11-09,11-10,11-11,11-12,11-13&series=Tokyo%20PM2.5:2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=WHO%2024%EC%8B%9C%EA%B0%84:15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=New+Delhi+PM2.5+%EC%B6%94%EC%9D%B4+%C2%B7+%EC%B5%9C%EA%B7%BC+30d+%C2%B7+%ED%98%84%EC%A7%80+%EC%8B%9C%EA%B0%81+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=WHO+24%EC%8B%9C%EA%B0%84%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## 특징

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

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The optional installation example uses an unversioned npm CLI, so following it can execute a newly published, unreviewed release with the user's privileges. This is a conditional supply-chain exposure rather than evidence of malicious code; the scanner incorrectly describes the installer as an MCP server.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The French installation example invokes an unversioned npm CLI, allowing future package releases to change the code executed during installation. A compromised release could execute with the installer's privileges, although this source shows neither an actual compromise nor an MCP server.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The Hindi README offers an installation command that does not pin the npm CLI version. Following this optional command can execute an unreviewed future release with local user privileges; this is a supply-chain weakness, not the MCP-server configuration claimed by the scanner.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The Japanese installation instructions reference an npm executable without an exact version, making the executed installer dependent on mutable package releases. A malicious or compromised future release could run with the user's privileges, but no actual compromise or MCP server is demonstrated.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The Korean README's optional npm installation command leaves the CLI version unspecified, exposing installation to unreviewed future package changes. Exploitation would require a malicious or compromised resolved release and could yield execution with user privileges; the MCP-server label is inaccurate.

Content

No source excerpt is available for this finding.

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.md (reported line 59)May include surrounding context.

md
0%C2%B7%20multi-city%20%C2%B7%20same%20parameter&unit=%C2%B5g%2Fm%C2%B3&labels=11-06,11-07,11-08,11-09,11-10,11-11,11-12,11-13&series=Tokyo%20PM2.5:5.14,2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:9.46,7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:37.3,58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=WHO%2024h:15,15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=New+Delhi+PM2.5+trend+%C2%B7+last+30d+%C2%B7+local+time+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-14%2C10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A104%2C73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=WHO+24h%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## Features

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · README.zh.md (reported line 59)May include surrounding context.

md
Fm%C2%B3&labels=11-06,11-07,11-08,11-09,11-10,11-11,11-12,11-13&series=Tokyo%20PM2.5:5.14,2.73,2.95,5,5.5,6.45,10.4,12.7&series=Seoul%20PM2.5:9.46,7.96,9.58,10.3,16.2,22.8,28.8,28.3&series=London%20PM2.5:37.3,58.3,82.4,49.4,23.9,15.8,10.8,13.7&series=WHO%2024h:15,15,15,15,15,15,15,15&w=960&h=500&scheme=dark&t=opendata
[sc6]: https://x-cmd.com/draw?type=line&title=New+Delhi+PM2.5+%E8%B6%8B%E5%8A%BF+%C2%B7+%E6%9C%80%E8%BF%91+30d+%C2%B7+%E7%AB%99%E7%82%B9%E5%BD%93%E5%9C%B0%E6%97%B6%E9%97%B4+%28Asia%2FKolkata%29&unit=%C2%B5g%2Fm%C2%B3&labels=10-14%2C10-15%2C10-16%2C10-17%2C10-18%2C10-19%2C10-20%2C10-21%2C10-22%2C10-23%2C10-27%2C10-28%2C10-29%2C10-30%2C10-31%2C11-01%2C11-02%2C11-03%2C11-04%2C11-05%2C11-06%2C11-07%2C11-08%2C11-09%2C11-10%2C11-11%2C11-12%2C11-13&series=New+Delhi+PM2.5%3A104%2C73.6%2C65.7%2C94%2C127%2C94%2C125%2C142%2C136%2C142%2C138%2C131%2C95%2C44.7%2C110%2C205%2C217%2C241%2C265%2C305%2C306%2C339%2C302%2C222%2C163%2C147%2C232%2C172&series=WHO+24h%3A15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15%2C15&w=960&h=500&t=opendata

## 特性

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

Medium
Category
MCP Rug Pull
Confidence
90% confidence
Finding

The optional installation command uses an unpinned npm executable and an unpinned skill repository, allowing their resolved contents to change between installations. Although this is not an MCP server reference as labeled, a compromised upstream release could execute code with the installer's permissions or install malicious 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
90% confidence
Finding

The optional installation command resolves an unpinned npm executable and an unpinned skill repository. This is an installation supply-chain risk rather than an MCP configuration issue: compromised upstream content could run with the installer's permissions or introduce malicious instructions into the agent's skills directory.

Content

No source excerpt is available for this finding.

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

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 · scripts/openaq_client.sh (reported line 7)May include surrounding context.

sh
# 依赖: curl / jq / awk(即本仓库零重型依赖主张)
set -euo pipefail

OPENAQ_API_BASE="${OPENAQ_API_BASE:-https://api.openaq.org/v3}"
AIR_CHART_CACHE_DIR="${AIR_CHART_CACHE_DIR:-$HOME/.cache/air-chart}"
AIR_CHART_TIMEOUT="${AIR_CHART_TIMEOUT:-20}"
AIR_CHART_MAX_RETRY="${AIR_CHART_MAX_RETRY:-4}"

The comment claims time-slot alignment, but series are joined by row position while their timestamps are discarded.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The renderer takes timestamp labels only from the first series and assigns all other measurements by row index, discarding their timestamps. If independently collected series have different missing intervals or start times, measurements are silently displayed at incorrect times, undermining city or pollutant comparisons and potentially misleading air-quality decisions.

Content

No source excerpt is available for this finding.