Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/shyftlabs/continuum/bottleneck-detectgit clone --depth 1 https://github.com/shyftlabs/continuumWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/shyftlabs/continuum/bottleneck-detect)<a href="https://agentmods.dev/commands/shyftlabs/continuum/bottleneck-detect"><img src="https://agentmods.dev/badge/commands/shyftlabs/continuum/bottleneck-detect.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00874 |
| Opus 5 | $0.00000 | $0.00437 |
| Sonnet 5 | $0.00000 | $0.00175 |
| Haiku 4.5 | $0.00000 | $0.00087 |
Grade A, and why
bottleneck-detect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
100% identical to bottleneck-detect — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
bottleneck detect
Analyze performance bottlenecks in swarm operations and suggest optimizations.
Usage
npx claude-flow bottleneck detect [options]
Options
--swarm-id, -s <id>- Analyze specific swarm (default: current)--time-range, -t <range>- Analysis period: 1h, 24h, 7d, all (default: 1h)--threshold <percent>- Bottleneck threshold percentage (default: 20)--export, -e <file>- Export analysis to file--fix- Apply automatic optimizations
Examples
Basic bottleneck detection
npx claude-flow bottleneck detect
Analyze specific swarm
npx claude-flow bottleneck detect --swarm-id swarm-123
Last 24 hours with export
npx claude-flow bottleneck detect -t 24h -e bottlenecks.json
Auto-fix detected issues
npx claude-flow bottleneck detect --fix --threshold 15
Metrics Analyzed
Communication Bottlenecks
- Message queue delays
- Agent response times
- Coordination overhead
- Memory access patterns
Processing Bottlenecks
- Task completion times
- Agent utilization rates
- Parallel execution efficiency
- Resource contention
Memory Bottlenecks
- Cache hit rates
- Memory access patterns
- Storage I/O performance
- Neural pattern loading
Network Bottlenecks
- API call latency
- MCP communication delays
- External service timeouts
- Concurrent request limits
Output Format
🔍 Bottleneck Analysis Report
━━━━━━━━━━━━━━━━━━━━━━━━━━━
📊 Summary
├── Time Range: Last 1 hour
├── Agents Analyzed: 6
├── Tasks Processed: 42
└── Critical Issues: 2
🚨 Critical Bottlenecks
1. Agent Communication (35% impact)
└── coordinator → coder-1 messages delayed by 2.3s avg
2. Memory Access (28% impact)
└── Neural pattern loading taking 1.8s per access
⚠️ Warning Bottlenecks
1. Task Queue (18% impact)
└── 5 tasks waiting > 10s for assignment
💡 Recommendations
1. Switch to hierarchical topology (est. 40% improvement)
2. Enable memory caching (est. 25% improvement)
3. Increase agent concurrency to 8 (est. 20% improvement)
✅ Quick Fixes Available
Run with --fix to apply:
- Enable smart caching
- Optimize message routing
- Adjust agent priorities
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 163 lines · 0 tokens per session scan A 2873e9bdeb14
bottleneck-detect is a command published in the GitHub repository shyftlabs/continuum (84 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 874 tokens. A static security scan graded it A with 0 findings. It is 100% identical to bottleneck-detect, differing in 0 lines, and is treated as a copy.
Other commands, from other repositories
end-to-end-run
Run one real objective through the whole orchestration loop as an operator, through the dashboard only, recording every collapse before fixing any of it.
pre-pr-review
Pre-PR review pipeline (checks + review agents + fixes + create PR).
aurelio-review-pr
Full PR review pipeline (local agents + external feedback + triage + fixes).
worktree
Manage parallel worktrees (setup, cleanup, status, tree, rebase).
setup-live-iterative
Stand up the local dev arm (backend from this worktree, in the running stack) plus a live-reload frontend.
new-setting
Add a settings knob end to end (definition, precedence, hot-reload, visibility).