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 skills add zystem-io/zymtrace-skills --skill configure-zymtrace-mcpgit clone --depth 1 https://github.com/zystem-io/zymtrace-skillsWrote 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/skills/zystem-io/zymtrace-skills/configure-zymtrace-mcp)<a href="https://agentmods.dev/skills/zystem-io/zymtrace-skills/configure-zymtrace-mcp"><img src="https://agentmods.dev/badge/skills/zystem-io/zymtrace-skills/configure-zymtrace-mcp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zystem-io/zymtrace-skills/configure-zymtrace-mcp"><img src="https://agentmods.dev/badge/skills/zystem-io/zymtrace-skills/configure-zymtrace-mcp.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00192 | $0.03523 |
| Opus 5 | $0.00096 | $0.01761 |
| Sonnet 5 | $0.00038 | $0.00705 |
| Haiku 4.5 | $0.00019 | $0.00352 |
Grade C, and why
configure-zymtrace-mcp scanned grade C with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Codex stores it as a `streamable_http` server in `~/.codex/config.toml`. Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
- **Never** inline the MCP token in the conversation, values files, scripts, commits, `mcp.json`, or any MCP-add command arguments. Always reference an env var (`$ZYMTRACE_MCP_TOKEN`, or `${env:ZYMTRACE_MCP_TOKEN}` in Cu Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsI <gateway-url>/health 2>&1 | head -3 How it starts
The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Configure zymtrace MCP
Helps the user connect their coding agent — Claude Code, OpenAI Codex, or Cursor — to their zymtrace backend's MCP server. Once connected, the user can analyze their CPU and GPU flamegraphs with natural-language queries — see optimize-cpu-workloads (CPU-only) or optimize-gpu-workloads (GPU) for the analytical workflow.
The MCP server is part of the zymtrace backend itself — it lives at
<gateway-host>/mcp. If the backend isn't installed and reachable yet, route toinstall-zymtrace-backendandexpose-zymtrace-backendfirst.
Which client? The URL, token, and reachability steps are identical for every client; only the add-the-server command differs (Step 3–4). Detect the client from context — which agent is running this, or what the user says ("connect Codex", "Cursor") — and use that client's block. If unsure, ask: "Claude Code, Codex, or Cursor?"
Greet the user
👋 Let's connect your agent to the zymtrace MCP so you can analyze flamegraphs in natural language. I'll need your zymtrace URL and — if your install has service-token auth on — a token. Five minutes max.
Doc reference: https://docs.zymtrace.com/mcp/configure-mcp
Skip the greeting if the user has already volunteered the URL and token.
Sources of truth
- MCP configuration: https://docs.zymtrace.com/mcp/configure-mcp
- Service tokens (auth): https://docs.zymtrace.com/authentication/service-tokens
- MCP overview: https://docs.zymtrace.com/mcp
Pre-flight
Check whether zymtrace is already connected in the user's client:
| Client | Check |
|---|---|
| Claude Code | claude mcp list 2>&1 | head -20 (and claude --version — older than v2.x: MCP support varies, recommend updating) |
| Codex | codex mcp list 2>&1 | head -20 |
| Cursor | open ~/.cursor/mcp.json (or project .cursor/mcp.json); or Settings → MCP/Tools |
If zymtrace already appears, the connection exists — jump straight to verification (Step 5).
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.
- 12d ago First seen · 230 lines · 192 tokens per session scan C 2194ba07bd3e
configure-zymtrace-mcp is a skill published in the GitHub repository zystem-io/zymtrace-skills (5 stars, last pushed 29d ago), licensed Apache-2.0. It adds 192 tokens to every session and 3,523 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it C with 3 findings (reads agent configuration directories, reads mcp configuration, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
skypilot-multi-cloud-orchestration
Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.
colab-finetuning
Fine-tune LLMs on Google Colab GPUs directly from openscience. Connects to Colab runtimes via WebSocket bridge for remote training with Unsloth. Supports SFT, GRPO, DPO, vision, and TTS workflows on free T4 to Pro A100 GPUs.
modal-serverless-gpu
Run approved CPU or GPU work through OpenScience computejob on the user's configured Modal account. Use for isolated scientific scripts, dependency provisioning, durable outputs, logs, status, cancellation, and recovery. Never invoke the Modal SDK or CLI directly.
review
Structured PR review for pygraphistry. Input: PR number/branch (default current branch PR). Output: findings and convergence artifacts under plans/ /. Method: multi-wave, evidence-first review across spec, correctness, tests, security, code quality, DRY, concurrency, performance, architecture, operability, and…
test-amplification
Methodical bug-driven test amplification for pygraphistry features. Use when hardening a feature area via user-workflow exploration, 5-Whys retrospectives, bug-taxonomy derivation, concrete test planning, implementation, and safety-gated validation.
lepton-cli
Operate NVIDIA DGX Cloud Lepton through the globally installed lep CLI. Use when the user asks to inspect or manage Lepton workspaces, endpoints/deployments, dev pods, batch jobs, fine-tuning jobs, Ray clusters, Slurm clusters, Dynamo endpoints, storage, secrets, nodes, ingress, templates, logs, authentication, or…