Claude Code Glyph
Score-based assessment (medium risk, 44/100). An AI review is available in the AI review tab.
Analysis record
- Analysed
- 3 days ago 19 November 2026
- Version
- v1.0.0
- Artifact
- SHA256 7837c3ba…bf2fd8
- Detail
- Full list
Is Claude Code Glyph safe?
Claude Code Glyph is a JetBrains IDE plugin listed with no description. It declares no special permissions at all: both the permission list and the host permission list are empty. The only network-style strings in the package are about twenty domain names lifted out of bundled files, including znny.ru and kotlinx-serialization-common.pro.
Most of those are not domains. One finding, XIOC-DOMAIN-kotlinx-serialization-common.pro, is the dotted filename of a Kotlin serialization library that the plugin bundles. The rest, cɟ.sj, g.ss, lr.ch, m.ai, t.ni, are two and three letter fragments of code identifiers that the scanner read as hostnames. Two of them contain characters no DNS server would resolve. None of them sits next to a line of code that reads a file and sends it somewhere.
The same pattern explains the other matches. The remaining items are low-severity code-quality rules that fire on ordinary JavaScript and Kotlin, and they name no specific capability, file, or destination. Nothing in this package runs a shell command on install, and nothing reaches for .env files, SSH keys, or cloud credentials. That last point matters most, because credential theft always leaves a path or a key name behind, and there is none here.
What the scanner tripped on is bundled library names and short code identifiers that look like addresses when you strip the surrounding code away. That is a limitation of the extractor, not something the plugin does. The empty permission list is the part worth trusting here.
Evidence ledger
Ranked by severity · findings with a source location link to the code viewer
YARA Rule Matches
8 rules| Severity | Rule | Hits | Files | Metadata |
|---|---|---|---|---|
| LOW | postinstall file download | 1 | ccglyph/lib/pty4j-0.13.12.jar | - |
| LOW | PM Zip with js | 1 | ccglyph/lib/ccglyph-1.0.0.jar | - |
| LOW | postinstall persistence mechanism | 3 | ccglyph/lib/ccglyph-1.0.0.jarccglyph/lib/annotations-24.0.1.jarccglyph/lib/pty4j-0.13.12.jar | - |
| LOW | postinstall crypto operations | 2 | ccglyph/lib/kotlin-stdlib-2.4.0.jarccglyph/lib/kotlinx-serialization-core-jvm-1.7.3.jar | - |
| LOW | JavaDropper | 1 | ccglyph/lib/kotlin-stdlib-2.4.0.jar | - |
| LOW | postinstall obfuscation | 4 | ccglyph/lib/kotlinx-serialization-json-jvm-1.7.3.jarccglyph/lib/kotlin-stdlib-2.4.0.jarccglyph/lib/kotlinx-serialization-core-jvm-1.7.3.jar +1 more | - |
| LOW | postinstall network communication | 1 | ccglyph/lib/pty4j-0.13.12.jar | - |
| LOW | postinstall system command | 5 | ccglyph/lib/ccglyph-1.0.0.jarccglyph/lib/slf4j-api-2.0.13.jarccglyph/lib/annotations-24.0.1.jar +2 more | - |
Publisher Evidence
Limited evidenceduckyman
Publisher identity, store signals, distribution reach, and warning signals used for context. Treat this as supporting evidence, not a clean bill of health.
12 evidence rows available.
Finding Categories
YARA Rules Matched
8 rules(18 hits)AI Security Report
AI Security Review
Evidence context: threat category none; evidence quality weak.
The supplied findings do not establish what behavior triggered the matches. A fresh analysis with the relevant code and surrounding context is needed. These findings neither establish malicious behavior nor demonstrate that the extension is safe.
Key Reasons
- Insufficient implementation context for the supplied findings
Reviewed 2026-11-14 · version 1.0.0; recommended action: reanalyze; model confidence 97%.
About This Extension
Frequently Asked Questions
Similar Extensions
Related extensions from the same publisher or marketplace
TermGlyph
duckyman
TASKING winIDEA
TASKING
ZenStack Language Tools
ZenStack
Thunder Client - REST API Client
thunder-client
C++ Modules Language Server
Sunrisepeak
Fork of the Zowe COBOL Language Support
29e596ce-1ecb-4683-ba4e-21007b4e93e4