Charles MCP Server
Related Servers
Alternatives to Charles MCP Server
No user-submitted related servers found.
Related Servers
- AlicenseAqualityCmaintenanceEnables AI agents to interact with Charles Proxy for HTTP traffic analysis, supporting incremental harvest, checkpoint timelines, and rich filtering tools.138MIT
- AlicenseBqualityBmaintenanceIntegrates Charles Proxy with MCP clients to provide real-time and historical network traffic capture and structured analysis. It features a summary-first approach that filters noise and desensitizes data for efficient, low-token agent debugging and monitoring.1876 PyPI317MIT
- AlicenseCqualityAmaintenanceTransforms mitmproxy into a toolset for AI agents to inspect, modify, and replay HTTP/HTTPS traffic in real-time.25146 PyPI125MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI to analyze real-time HTTP(S) traffic captured by ProxyPin, with tools to browse, search, and inspect requests and responses, as well as access saved history sessions.9 npm4MIT
- AlicenseBqualityCmaintenanceEnables AI agents to intercept, inspect, modify, and replay HTTP, WebSocket, TCP, and UDP traffic through mitmproxy, with controlled security constraints and process isolation.18AGPL 3.0
- AlicenseNot gradedqualityCmaintenanceEnables MCP-compatible AI assistants to directly search and inspect Whistle-captured HTTP, HTTPS, and WebSocket traffic, including full session details, timing, and frame analysis, without modifying Whistle configuration.4 npmMIT
TDQS
Scored across 47 tools
Many tools have intentionally duplicated surfaces across live/history/reverse planes (read_live_capture vs reverse_read_live_entries, query_recorded_traffic vs analyze_recorded_traffic vs reverse_query_entries, mock_* vs mock_rule_*), and descriptions carry most of the burden. Prefixes like mock_ and reverse_ help, but an agent could easily select the wrong capture/read/query variant without reading deeply.
Most tools follow a consistent [domain]_[verb]_[object] pattern (start_live_capture, mock_get, reverse_replay_entry), and the mock_/reverse_ prefixes make families recognizable. Minor deviations like throttling, reset_environment, charles_status, and mock_dispatcher break the verb-first pattern but do not obscure meaning.
47 tools is well into the 'too many' range and reflects substantial duplication: several near-parallel subsystems (live/history/reverse/mock-rule) each re-implement list/read/query/stop-style operations. The broad Charles domain justifies many tools, but consolidation would meaningfully improve the surface.
The surface is broad and covers the main lifecycle: live capture, saved recordings, filtering/detail drill-down, mocking, replay, reverse analysis, and environment control. Minor gaps such as no way to start/quit Charles or delete recordings/imports can be worked around, but they are not fatal dead ends.