"author:0xmortuex" matching MCP servers:

Enables an AI agent to parse project lockfiles (npm, pnpm, Yarn, uv, Poetry, requirements.txt, Cargo, go.sum) and query OSV.dev to find known vulnerabilities in locked versions, transitive dependencies included, then produce minimal upgrade plans with breaking-change flags, cleanliness re-checks, and the parent dependency responsible. It also lists outdated packages and explains individual advisories, all read-only with no API key required.

Enables developers to query a project's Python and JavaScript/TypeScript import graph for dependents, blast radius, affected tests and entry points, cycles, and dead-code candidates, with a confidence summary on every answer.

Enables agents to diagnose environment-variable issues read-only without exposing values, including .env versus template comparison, undocumented or unused variables across source code, shape validation, secret-exposure checks, git-ignore and history checks, and safe .env.example proposals.

Enables AI agents to analyze browser HAR network traces for slow requests with per-phase timing, failures, waterfalls, and critical paths. Masks secrets by default and can write sanitized copies for sharing.

Enables an AI agent to compute path-level semantic differences between two JSON, YAML, TOML, CSV, .env or INI files or pasted strings — matching lists by key so moves aren't reported as edits, honoring numeric tolerance and ignore patterns, and returning verified RFC 6902 patches. It also streams keyed CSV row diffs and inspects how a file parses, all read-only within allowed folders.

Enables AI agents to inspect, list, read, search, verify, create, and safely extract zip/tar archives, including reading members without extraction, while blocking path traversal, symlink escapes, and decompression bombs.

Enables AI agents to safely run test suites over MCP, returning concise JUnit-based results, trimmed tracebacks, rerun-failed support, and flaky-test detection while handling timeouts, process cleanup, and bounded output.

Enables AI agents to run the same regex in real Python, JavaScript, and optional regex engines side by side, flag disagreements, confirm ReDoS by timing adversarial input in a killed child process, explain patterns, convert Python/JavaScript syntax, preview substitutions, and generate verified test suggestions. All execution is isolated in throwaway processes with hard timeouts and strict input caps.

Enables AI agents to run a project's pytest, unittest, jest, or vitest suites and get back compact structured results with trimmed tracebacks, collection errors kept apart from real failures, and the slowest tests. Agents can then re-run just the failures, expand one failure with local variables, repeatedly execute a single test to judge flakiness, and inspect Python coverage gaps, all under a hard timeout that kills the whole process tree.

Enables an AI agent to inspect which process is listening on a local port and safely plan, confirm, and verify termination of that process across platforms.

Enables an agent to mine a repository's git history and answer questions a new teammate would ask: which files are riskiest based on churn, recency and author breadth, who actually knows each file based on surviving lines, which files secretly change together, which directories depend on one person, and what recent commits mean for release notes. All analysis runs through a read-only server sandboxed to a single directory.

Enables AI agents to investigate very large log files with constant-memory streaming, offering format overviews, filtered searches, error clustering, timelines, request tracing, and window comparisons across plain, gzipped, and rotated logs.

Enables AI agents to identify which processes are using ports across Windows, macOS, and Linux, inspect process details and trees, find free ports, test TCP reachability, and safely free ports through a two-step confirmed kill that refuses dangerous targets.

Enables AI assistants to mine a git repository's history and answer pre-change questions: which files are riskiest, who genuinely knows a file or directory (blame-weighted, with bus factor and inactive authors), which files secretly change together, churn timelines, commit search, and release notes. It runs read-only against real git with noise filtering and reproducible, documented formulas.

Enables an AI agent to analyze browser HAR exports, break down request timing, identify dominant phases and critical request chains, render ASCII waterfalls, and safely redact secrets before analysis.

Enables AI agents to scan a working tree and git history for committed credentials without exposing raw secrets, returning redacted findings and provider-specific rotation, misuse-check, and clean-up guidance.