Agentic AI Atlasby a5c.ai
OverviewWikiGraphFor AgentsEdgesSearchWorkspace
/
GitHubDocsDiscord
iiRecord
Agentic AI Atlas · Additional Reference Projects
page:docs-agent-mux-archive-comparisons-17-additional-referencesa5c.ai
Search record views/
Record · tabs

Available views

II.Record viewspp. 1 - 1
overviewarticlejsongraph
II.
Page JSON

page:docs-agent-mux-archive-comparisons-17-additional-references

Structured · live

Additional Reference Projects json

Inspect the normalized record payload exactly as the atlas UI reads it.

File · wiki/docs/agent-mux/archive/comparisons/17-additional-references.mdCluster · wiki
Record JSON
{
  "id": "page:docs-agent-mux-archive-comparisons-17-additional-references",
  "_kind": "Page",
  "_file": "wiki/docs/agent-mux/archive/comparisons/17-additional-references.md",
  "_cluster": "wiki",
  "attributes": {
    "nodeKind": "Page",
    "sourcePath": "docs/agent-mux/archive/comparisons/17-additional-references.md",
    "sourceKind": "repo-docs",
    "title": "Additional Reference Projects",
    "displayName": "Additional Reference Projects",
    "slug": "docs/agent-mux/archive/comparisons/17-additional-references",
    "articlePath": "wiki/docs/agent-mux/archive/comparisons/17-additional-references.md",
    "article": "\n# Additional Reference Projects\n\n> Archived comparison document. Preserved for historical context; not part of the current normative `reference/` contract.\n\nExtension of `16-reference-comparison.md`. Candidates surfaced via WebSearch on 2026-04-12 across four categories:\n\n1. Monitoring/orchestration CLIs wrapping coding-agent CLIs\n2. UI wrappers over agent harnesses\n3. Multi-harness SDKs/runners\n4. Session/token analytics\n\n## Candidates\n\n| # | Name | URL | Stars (approx) | Language | Category | Agents supported | Relevance to agent-mux |\n|---|---|---|---|---|---|---|---|\n| 1 | cctop | https://github.com/st0012/cctop | ~1k | Swift | Monitoring | claude-code, opencode | Menubar switcher between sessions — UX pattern for our `sessions` surface. |\n| 2 | ccmonitor | https://github.com/shinagaki/ccmonitor | ~500 | TS/Node | Monitoring | claude-code | 5-hour rolling window reporting we don't model in `CostRecord`. |\n| 3 | cc-monitor-rs | https://github.com/ZhangHanDong/cc-monitor-rs | ~300 | Rust (Makepad) | Monitoring | claude-code | Native real-time UI consuming JSONL — validates our event stream shape. |\n| 4 | Claude-Code-Usage-Monitor | https://github.com/Maciek-roboblog/Claude-Code-Usage-Monitor | ~6k | Python | Analytics | claude-code | ML-based burn-rate prediction — gap in agent-mux. |\n| 5 | ccboard | https://github.com/FlorianBruniaux/ccboard | ~400 | Rust | Monitoring + Dashboard | claude-code | Unifies sessions, costs, hooks, agents, MCP in one binary — mirrors our CLI surface. |\n| 6 | cc-viewer | https://github.com/weiesky/cc-viewer | ~200 | TS | Monitoring | claude-code | Live request/response capture for debugging — could inspire a `--trace` mode. |\n| 7 | claude-code-monitor (zcquant) | https://github.com/zcquant/claude-code-monitor | ~150 | TS | Analytics | claude-code | OpenTelemetry OTLP dashboard — telemetry export gap. |\n| 8 | ccusage | https://github.com/ryoppippi/ccusage | ~20k | TS | Analytics | claude, codex, opencode, pi, amp | JSONL session reader family — already tracked; listed for completeness. |\n| 9 | claude-usage (phuryn) | https://github.com/phuryn/claude-usage | ~200 | TS | Analytics | claude-code | Pro/Max subscription progress-bar UX. |\n| 10 | vibe-kanban | https://github.com/BloopAI/vibe-kanban | ~15k | Rust | UI wrapper | claude, copilot, cursor, codex, gemini, qwen, droid, opencode, amp, ccr | Kanban + per-agent worktree — gap: we lack a default UI. |\n| 11 | paperclip | https://github.com/paperclipai/paperclip | ~30k | TS/React | SDK + UI | claude, codex, cursor, openclaw, pi, bash, http | Already tracked; closest structural peer. |\n| 12 | claude-squad | https://github.com/smtg-ai/claude-squad | ~7k | Go | Terminal orchestrator | claude, codex, gemini, aider, opencode, amp | tmux-based parallel workspaces with yolo mode — gap. |\n| 13 | aider | https://github.com/Aider-AI/aider | ~42k | Python | Agent harness | multi-LLM (Claude, o1, DeepSeek, local) | Tree-sitter repo map + architect mode — we don't ship these. |\n| 14 | continue | https://github.com/continuedev/continue | ~26k | TS | IDE wrapper | any (Claude, GPT, local, Mistral) | VS Code/JetBrains integration — agent-mux has no IDE surface. |\n| 15 | goose (block) | https://github.com/block/goose | ~29k | Rust | Agent framework | 15+ providers via MCP | 3000+ MCP tools + recipes — our MCP manager is narrower. |\n| 17 | OpenHarness (HKUDS) | https://github.com/HKUDS/OpenHarness | ~400 | Python | Harness | Claude, OpenAI, Copilot, Codex, Kimi, GLM, MiniMax | Academic harness with pluggable endpoints — reference for capability surface. |\n\n## Cross-reference: features agent-mux lacks\n\n| Feature | Covered by | Priority |\n|---|---|---|\n| Burn-rate / limit prediction (5-hour rolling) | Claude-Code-Usage-Monitor, ccmonitor | High — pairs with existing `CostRecord`. |\n| OpenTelemetry / OTLP export | claude-code-monitor (zcquant), ccboard | Medium — we have events but no OTLP exporter. |\n| Built-in Kanban / task UI | vibe-kanban, paperclip | Medium — scope decision (SDK-first vs UI-first). |\n| Tree-sitter repo map / architect planning | aider | Medium — out of scope for pure harness, but useful as plugin. |\n| IDE extension (VS Code / JetBrains) | continue | Low — different shape of product. |\n| MCP recipe library (3000+ tools) | goose | Medium — we ship MCP lifecycle but no curated catalog. |\n| Live request/response trace viewer | cc-viewer | Low — debugging aid. |\n| Menubar / system tray switcher | cctop | Low — presentation layer only. |\n| Qwen / droid / amp adapters | vibe-kanban, ccusage, claude-squad | Medium — already noted in `16-reference-comparison.md`. |\n\n## Net assessment\n\nAdding these 17 projects to the survey in `16-reference-comparison.md` does not change the top-line conclusion: **agent-mux remains a structural superset on the SDK surface**. However three previously-unflagged gaps emerge:\n\n1. **Predictive burn-rate analytics** (Claude-Code-Usage-Monitor) — additive to `CostRecord`.\n2. **OTLP telemetry exporter** (zcquant, ccboard) — natural plugin.\n3. **MCP recipe catalog** (goose) — complements our MCP lifecycle manager.\n\nThe UI-shaped projects (vibe-kanban, paperclip, continue) remain orthogonal: agent-mux is an SDK, not a UI, and should stay that way per scope doc.\n",
    "documents": []
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "page:docs-agent-mux-archive-comparisons",
      "to": "page:docs-agent-mux-archive-comparisons-17-additional-references",
      "kind": "contains_page"
    }
  ]
}

Shortcuts

Back to overview
Open graph tab