: {\n> 65 | \t\turl: \"http://localhost:11434\"\n 66 | \t}\n 67 | };\n 68 | ","networkType":"active-outbound","sendsData":true,"dataDescription":"URL property targets localhost (http://localhost:11434) (allowlisted)","endpointSource":"hardcoded-url"},{"host":"raw.githubusercontent.com","url":"https://raw.githubusercontent.com/aldo-g/obsidian-llm-test/master/community-models.json","filePath":"src/services/llm.ts","lineNumber":742,"codeSnippet":" 739 | ): Promise> {\n 740 | try {\n 741 | const url = `https://raw.githubusercontent.com/aldo-g/obsidian-llm-test/master/community-models.json`;\n> 742 | const response = await requestUrl({ url });\n 743 | \n 744 | if (response.status === 200) {\n 745 | const data = response.json;","networkType":"active-inbound","sendsData":false,"dataDescription":null,"endpointSource":"hardcoded-url"}],"initialDiff":null,"initialDiffFrom":"1.1.0","initialDiffTo":"1.2.2","summary":"This plugin reads all markdown file contents from your Obsidian vault and sends them to external LLM providers—OpenAI, Anthropic, DeepSeek, Google, and Mistral—to generate AI-powered test questions based on your notes. It can also connect to local AI services as an alternative. The plugin retrieves available models from these cloud providers to let you choose which LLM to use for test generation. No data obfuscation is applied before transmission.","dependencyVulnerabilities":[{"packageName":"ajv","packageVersion":"<6.14.0","severity":"moderate","title":"ajv has ReDoS when using `$data` option","advisoryUrl":"https://github.com/advisories/GHSA-2g4f-4pwh-qvx6","ghsaId":"GHSA-2g4f-4pwh-qvx6","cveId":null,"vulnerableRange":"<6.14.0","fixAvailable":null,"source":"npm-audit","isDirect":false,"path":null},{"packageName":"brace-expansion","packageVersion":"<1.1.13","severity":"moderate","title":"brace-expansion: Zero-step sequence causes process hang and memory exhaustion","advisoryUrl":"https://github.com/advisories/GHSA-f886-m6hf-6m8v","ghsaId":"GHSA-f886-m6hf-6m8v","cveId":null,"vulnerableRange":"<1.1.13","fixAvailable":null,"source":"npm-audit","isDirect":false,"path":null},{"packageName":"esbuild","packageVersion":"<=0.24.2","severity":"moderate","title":"esbuild enables any website to send any requests to the development server and read the response","advisoryUrl":"https://github.com/advisories/GHSA-67mh-4wv8-2f99","ghsaId":"GHSA-67mh-4wv8-2f99","cveId":null,"vulnerableRange":"<=0.24.2","fixAvailable":"0.28.0","source":"npm-audit","isDirect":true,"path":null},{"packageName":"flatted","packageVersion":"<3.4.0","severity":"high","title":"flatted vulnerable to unbounded recursion DoS in parse() revive phase","advisoryUrl":"https://github.com/advisories/GHSA-25h7-pfq9-p65f","ghsaId":"GHSA-25h7-pfq9-p65f","cveId":null,"vulnerableRange":"<3.4.0","fixAvailable":null,"source":"npm-audit","isDirect":false,"path":null},{"packageName":"flatted","packageVersion":"<=3.4.1","severity":"high","title":"Prototype Pollution via parse() in NodeJS flatted","advisoryUrl":"https://github.com/advisories/GHSA-rf6f-7fwh-wjgh","ghsaId":"GHSA-rf6f-7fwh-wjgh","cveId":null,"vulnerableRange":"<=3.4.1","fixAvailable":null,"source":"npm-audit","isDirect":false,"path":null},{"packageName":"minimatch","packageVersion":"<3.1.3","severity":"high","title":"minimatch has a ReDoS via repeated wildcards with non-matching literal in pattern","advisoryUrl":"https://github.com/advisories/GHSA-3ppc-4f35-3m26","ghsaId":"GHSA-3ppc-4f35-3m26","cveId":null,"vulnerableRange":"<3.1.3","fixAvailable":null,"source":"npm-audit","isDirect":false,"path":null},{"packageName":"minimatch","packageVersion":"<3.1.3","severity":"high","title":"minimatch has ReDoS: matchOne() combinatorial backtracking via multiple non-adjacent GLOBSTAR segments","advisoryUrl":"https://github.com/advisories/GHSA-7r86-cg39-jmmj","ghsaId":"GHSA-7r86-cg39-jmmj","cveId":null,"vulnerableRange":"<3.1.3","fixAvailable":null,"source":"npm-audit","isDirect":false,"path":null},{"packageName":"minimatch","packageVersion":"<3.1

LLM Test Generator