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.
/plugin marketplace add curie-eng/curienpx agentmods add plugins/curie-eng/curie/sre-botgit clone --depth 1 https://github.com/curie-eng/curieGrade A, and why
sre-bot 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 3d 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.
What it actually says
{
"name": "sre-bot",
"version": "0.1.0",
"description": "SRE triage assistant for plain-English production health and Kubernetes questions in Slack. Read-only by default: the Kubernetes read connector exposes only read tools, and its credential cannot read Secrets or mutate resources. Grafana and Tempo connectors add observability reads. The one declared write verb rolls a single allowlisted Deployment using a separate credential and the exact human approval gate mcp__k8s-write__restart_deployment. A second gated verb scales a single allowlisted Deployment through the `deployments/scale` subresource with its own credential and the gate mcp__k8s-scale__scale_deployment; it reports the replica count it read before writing, so the platform can offer an undo.",
"license": "Apache-2.0",
"author": {
"name": "CurieTech"
},
"keywords": [
"sre",
"observability",
"grafana",
"loki",
"prometheus",
"pyroscope",
"kubernetes",
"k8s",
"cluster",
"triage",
"read-only"
],
"starterPrompts": [
"Is anything broken right now?",
"Is any pod crashlooping or restarting?",
"Are all the nodes healthy?",
"Did anything get OOMKilled today?",
"Are there any alerts firing?",
"Which services are logging the most errors today?",
"Is the API slow right now?",
"Show me slow traces from the last hour"
],
"approvalPolicy": {
"gates": [
{
"gate": "mcp__k8s-write__restart_deployment",
"route": "sre-approvals"
},
{
"gate": "mcp__k8s-scale__scale_deployment",
"route": "sre-approvals"
},
{
"gate": "mcp__self-upgrade__upgrade_self",
"route": "sre-approvals"
}
]
}
}
What it installs
The manifest is a name and a version. 1 skill travel with it, and installing the plugin installs all of them — 209 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 3d ago First seen · 49 lines scan A 6a41be3b62bd
sre-bot is a plugin published in the GitHub repository curie-eng/curie (35 stars, last pushed 3d ago), licensed Apache-2.0. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other plugins, from other repositories
defang
Deploy Docker Compose apps to your own cloud account using Defang. Includes a /defang:deploy skill and Defang MCP server integration.
slicer marketplace
SlicerVM skills for Claude Code: Linux microVM workflows, git worktrees in VMs, filtered egress proxy, RustFS S3-compatible storage, and k3sup Kubernetes provisioning.
ci-incident-medic
Triages failing GitHub Actions runs, hardens Dockerfiles, pre-flights Kubernetes manifests before they reach the cluster, and drafts blameless postmortems from incident evidence.
parallel-lifecycle
Give every git worktree its own isolated, disposable environment (app port, ephemeral Chrome/CDP, per-worktree database) behind a single .env.worktree contract, so N agentic coding sessions build features in parallel without colliding on ports, the browser, or the database.
arcane-mcp-server
Docker infrastructure management with 180+ MCP tools, workflow guidance, and safety guardrails. Connects to the Arcane Docker Management platform.
chainguard-codegen
Generate secure Dockerfiles and apko configurations, and migrate existing Dockerfiles to use Chainguard Containers.