Page 42
A critical vulnerability in the Telegram OAuth provider allowed attackers to bypass authentication by submitting fabricated hash values without cryptographic validation. The fix replaces an unsafe string comparison with PHP's constant-time `hash_equals()` function, preventing timing attacks and ensuring all user data parameters are properly validated against the HMAC-SHA256 signature.
A high-severity security finding was identified in `templates/claude-workflow/workflows/claude.yml` where `secrets: inherit` passed every repository secret to a reusable workflow, violating the principle of least privilege. The fix explicitly passes only `CLAUDE_CODE_OAUTH_TOKEN`—the single secret the called workflow actually needs—drastically reducing the blast radius if the reusable workflow is ever compromised.
A critical integer underflow vulnerability was discovered in tree-sitter's array.h header file, where the `_array__splice()` function calculated array sizes without proper bounds checking. The vulnerable code relied on assert() statements that are disabled in release builds, allowing arithmetic underflow when `old_count > *size + new_count`, potentially causing memory corruption through out-of-bounds memcpy operations.
A critical Server-Side Request Forgery vulnerability in `popup.js` allowed attackers to inject malicious URLs from scraped webpages directly into fetch() calls, potentially accessing internal network resources and AWS metadata endpoints. The fix adds URL validation to ensure only HTTP/HTTPS protocols are used and blocks requests to private IP ranges and localhost addresses.
A high-severity vulnerability (CVE-2026-42043) in axios versions prior to 1.15.1 allowed attackers to bypass NO_PROXY environment variable restrictions using specially crafted URLs. This meant HTTP requests intended to stay internal could be routed through an attacker-controlled proxy, potentially exposing sensitive data. The fix upgrades axios to version 1.15.1, which correctly validates URLs against NO_PROXY rules.
A critical validation bypass in the `/api/agents/jobs` endpoint allowed attackers to send malformed JSON with arbitrary properties that could trigger prototype pollution or unexpected behavior. The fix added comprehensive validation checks including array detection and property existence verification to prevent malicious payloads from reaching downstream processing logic.
A critical Server-Side Request Forgery (SSRF) vulnerability was discovered in `hubcmdui/routes/httpProxy.js` where the `/proxy/test` endpoint passed a user-controlled `testUrl` parameter directly to `axios.get()` without any validation. An attacker could exploit this to probe internal infrastructure — including AWS metadata endpoints, Redis instances, and private network ranges. The fix introduces a strict URL validation function that blocks private IP ranges, loopback addresses, and non-HTTP(S)
The Model Context Protocol (MCP) TypeScript SDK in version 1.17.1 lacked DNS rebinding protection, leaving applications vulnerable to a sophisticated network-level attack. By upgrading to version 1.24.0, DNS rebinding protection is now enabled by default, preventing attackers from exploiting the SDK's HTTP request handling to access internal resources or bypass security controls.
A critical vulnerability in `js/config.js` allowed attackers to inject malicious configuration data through DNS spoofing or man-in-the-middle attacks. The application fetched remote JSON configuration from GitHub without validating the response structure, enabling arbitrary code execution through crafted payloads. A single validation check now ensures only properly-formed configuration objects are accepted.
A high-severity shell injection vulnerability was discovered in `.github/commaSplitter/action.yaml` where unsanitized user input was directly interpolated into a bash `run:` step using `${{ inputs.input }}`. An attacker could craft a malicious input string to escape the shell command and execute arbitrary code on the GitHub Actions runner, potentially stealing secrets and source code. The fix introduces an intermediate environment variable to safely pass the input without shell interpretation.
A Jekyll plugin used unsafe Ruby backticks to execute a `git log` command with an unescaped file path, creating a command injection vulnerability. By switching to `Open3.capture2()` with argument array syntax, the fix prevents shell interpretation and eliminates the attack surface entirely.
A missing batch size limit in `BatchModeRunner.jsx` allowed users to trigger unlimited LLM API calls by pasting thousands of items into the batch input field. This could exhaust shared API quotas in organizational settings where a single API key is distributed across multiple users. The fix introduces a hard cap of 25 items (`MAX_BATCH_SIZE = 25`) enforced directly in the `canRun()` validation function.