Local-first MCP server testing with the botz MCP Swarm CLI. Skills install the botz host and its mcp-swarm extension (pinned 0.15.0) on first use, run only when invoked by name, and wire up no MCP server by default; Botz platform mode is opt-in via /z-mcp-swarm-platform.
Plugin botz-mcp-swarm from marketplace zoto-agents in zotoio/zoto-agents. Commands read from the repo's manifest — 6 days ago.
mcp-swarm-analyseAnalyse a failed botz MCP Swarm run with the configured model, review its proposed suite fixes one by one, and accept or reject them by explicit proposal id. Nothing changes without a named id.
mcp-swarm-compareCompare two botz MCP Swarm runs or a series of runs and visualise latency, error rate, and tool-eval trends in Cursor's /canvas or a self-contained HTML page. Use to check for regressions between builds, environments, or load settings.
mcp-swarm-descriptionsReview an MCP server's tool descriptions, propose clearer ones, examine description suggestions saved by botz MCP Swarm optimize, and measure a description change with tool evals. Use when models pick the wrong tool, misuse arguments, or a server's descriptions need an audit.
mcp-swarm-diagnoseDiagnose failed botz MCP Swarm suites, missing credentials, invalid fixtures, execution errors, and performance regressions from CLI diagnostics and run reports.
mcp-swarm-flowTurn a task described in words, such as "find an open bug in GitHub, file it in Jira, tell #triage", into a botz MCP Swarm suite whose steps call several MCP servers in order and pass values between them. Use to load-test or regression-test an agent-like workflow that spans MCP servers.
mcp-swarm-guidedGuided, question-driven introspection of an MCP server with botz MCP Swarm, from target and credentials through suite generation, review, and execution to a summarised result. Use when someone wants to test an MCP server and has not set up a suite yet.
mcp-swarm-platformOpt in to Botz MCP Swarm platform mode and use its features (central swarm MCP server with OAuth, remote suite runs, worker dispatch, service suites, swarm_* MCP tools). Use only when the user explicitly enables platform mode or BOTZ_MCP_SWARM_MODE=platform is set.
mcp-swarm-record-replayRecord redacted MCP wire traffic from a server with botz MCP Swarm and replay it against the same or another server, as recorded, at maximum rate, or paced. Use to reproduce a session, compare server builds on identical traffic, or create fixtures.
mcp-swarm-reportVisualise one botz MCP Swarm run in Cursor's /canvas or a self-contained HTML page, with a tool selector showing each tool's latency, errors, and a sample request and response. Use after a run, or to explain a run report.
mcp-swarm-server-updatesCheck MCP servers under test for updates (serverInfo version, negotiated MCP spec/protocol version, capabilities, and every tool, resource, template and prompt schema), review the diff, surgically update only the suites and scenarios that reference changed primitives, and record a changelog entry on
mcp-swarm-tagsEdit descriptions on MCP Swarm suites and servers, and add ad-hoc tags to suites, scenarios and runs, then filter lists by tag from the local CLI.
mcp-swarm-upgradeUpgrade the botz MCP Swarm CLI and move saved suites, credentials, run reports, and service state into the current layout. Use when updating MCP Swarm, after release notes mention moved files, or when old .botz/suites, .runs, .swarm-data, or .otel directories are present.
mcp-swarmCapture, validate, refresh, and run MCP test suites entirely locally with the botz MCP Swarm CLI (no central Swarm server, no MCP server config). Use for testing MCP servers, repeatable tool assertions, local load tests, and CI reports.
/botz-mcp-swarm:z-mcp-swarm-analyseRun the mcp-swarm-analyse skill. Analyse a failed botz MCP Swarm run, review proposed suite fixes, and accept or reject them by explicit proposal id.
/botz-mcp-swarm:z-mcp-swarm-compareRun the mcp-swarm-compare skill. Compare two botz MCP Swarm runs or a series of runs and visualise latency, error rate, and tool-eval trends in Cursor's /canvas or a self-contained HTML page. Use to check for regressions between builds, environments, or load settings.
/botz-mcp-swarm:z-mcp-swarm-descriptionsRun the mcp-swarm-descriptions skill. Review an MCP server's tool descriptions, propose clearer ones, examine description suggestions saved by botz MCP Swarm optimize, and measure a description change with tool evals. Use when models pick the wrong tool, misuse arguments, or a server's descriptions
/botz-mcp-swarm:z-mcp-swarm-diagnoseRun the mcp-swarm-diagnose skill. Diagnose failed botz MCP Swarm suites, missing credentials, invalid fixtures, execution errors, and performance regressions from CLI diagnostics and run reports.
/botz-mcp-swarm:z-mcp-swarm-flowRun the mcp-swarm-flow skill. Turn a task described in words into a botz MCP Swarm suite whose steps call several MCP servers in order and pass values between them.
/botz-mcp-swarm:z-mcp-swarm-guidedRun the mcp-swarm-guided skill. Guided, question-driven introspection of an MCP server with botz MCP Swarm, from target and credentials through suite generation, review, and execution to a summarised result. Use when someone wants to test an MCP server and has not set up a suite yet.
/botz-mcp-swarm:z-mcp-swarm-platformRun the mcp-swarm-platform skill. Opt in to Botz MCP Swarm platform mode (central swarm MCP server, remote runs, workers) or turn it off.
/botz-mcp-swarm:z-mcp-swarm-record-replayRun the mcp-swarm-record-replay skill. Record redacted MCP wire traffic from a server with botz MCP Swarm and replay it against the same or another server, as recorded, at maximum rate, or paced. Use to reproduce a session, compare server builds on identical traffic, or create fixtures.
/botz-mcp-swarm:z-mcp-swarm-reportRun the mcp-swarm-report skill. Visualise one botz MCP Swarm run in Cursor's /canvas or a self-contained HTML page, with overall and per-tool VUs/TPS over time (loadTimeline/timeSeries) and a tool selector showing each tool's latency, errors, and sample request/response.
/botz-mcp-swarm:z-mcp-swarm-server-updatesRun the mcp-swarm-server-updates skill. Check MCP servers under test for updates (serverInfo version, negotiated MCP spec/protocol version, capabilities, and every tool, resource, template and prompt schema), review the diff, surgically update only the suites and scenarios that reference changed pri
/botz-mcp-swarm:z-mcp-swarm-tagsRun the mcp-swarm-tags skill. Edit descriptions on MCP Swarm suites and servers, and add ad-hoc tags to suites, scenarios and runs, then filter lists by tag from the local CLI.
/botz-mcp-swarm:z-mcp-swarm-upgradeRun the mcp-swarm-upgrade skill. Upgrade the botz MCP Swarm CLI and move saved suites, credentials, run reports, and service state into the current layout. Use when updating MCP Swarm, after release notes mention moved files, or when old .botz/suites, .runs, .swarm-data, or .otel directories are pre
/botz-mcp-swarm:z-mcp-swarmRun the mcp-swarm skill. Capture, validate, refresh, and run MCP test suites with the botz MCP Swarm CLI. Use for testing MCP servers, repeatable tool assertions, load tests, and CI reports.
Compress rules, code, images, and web pages into CRUX notation, with CRUX memories and meditation. Packages the CRUX-Compress release; run /z-crux-init once per project. Hooks need Python 3.
14 skills · 21 commands · 8 agents
Stamp static (pytest) and LLM (@cursor/sdk) eval suites side-by-side, detect drift when AI primitives change and update evals with user confirmation, compare runs across models on /canvas. Surgical update on git diff, sovereign user-authored cases.
9 skills · 13 commands · 8 agents
Decompose complex initiatives into reviewable specs, judge them independently, then execute with adversarial verification. Live execution status aggregation and no-restart token-budget configuration.
3 skills · 4 commands · 3 agents
Live updating htop-style CLI that shows every Cursor agent on this machine, with expandable parent-to-subagent hierarchy and scrolling log tails
1 skill · 2 commands · 1 agent
Consume governed Agent Skills and subagents from the hosted botz.ai MCP server (https://mcp.botz.ai/mcp, OAuth sign-in) with zero-drift guid+version pinning and CRUX-compressed downloads.
1 skill · 1 command · 1 MCP server
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.
3 skills · 6 commands · 1 MCP server · hooks
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
8 skills · 1 MCP server
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
Claude Code skill for creating and presenting Slidev slidedecks
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.