Security vulnerabilities and automated fixes for argument injection issues
3 posts found
A protocol replay-check CLI built its subprocess argument list by calling `str.format()` on a user-supplied `--command` template and then handing the result to `shlex.split()`, so a sample filename containing spaces, quotes, or shell metacharacters could split into extra argv entries — or execute as shell code when the template wrapped the placeholder in `sh -c`. The fix wraps the interpolated path in `shlex.quote()` before formatting, so the path always survives `shlex.split()` as a single toke
A critical unauthenticated remote code execution vulnerability in Crawl4AI 0.8.9 allowed attackers to inject arbitrary Chromium launch arguments through the `browser_config.extra_args` parameter, potentially taking full control of the host process. The fix upgrades to Crawl4AI 0.9.0 and refactors the crawler initialization in `agent/tools/crawler.py` to use the new `BrowserConfig` and `CrawlerRunConfig` APIs, which enforce proper argument validation. This change eliminates the injection surface
A high-severity argument injection vulnerability was discovered in the Copilot tool bridge (`bridge.ts`) where user-controlled `request.args` were passed directly to `tool.execute()` without any validation or sanitization. The fix introduces Zod schema validation at line 108, ensuring that tool arguments are parsed against a declared `inputSchema` before execution. This prevents malformed or malicious payloads — including prototype pollution attempts — from reaching the underlying tool implement