microsoft/sre-agent is the community repository for Azure SRE Agent, an AI assistant for diagnosing and resolving production problems and reducing routine operations work. It is for teams working on service reliability and includes documentation, discussions, videos, and hands-on labs for using the product. The catalogue skills relate to the agent's official plugins and resources.
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 microsoft/sre-agent --skill rbac-auditgit clone --depth 1 https://github.com/microsoft/sre-agentWrote 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/microsoft/sre-agent/rbac-audit)<a href="https://agentmods.dev/skills/microsoft/sre-agent/rbac-audit"><img src="https://agentmods.dev/badge/skills/microsoft/sre-agent/rbac-audit/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/microsoft/sre-agent/rbac-audit"><img src="https://agentmods.dev/badge/skills/microsoft/sre-agent/rbac-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 60 Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.Fix: Set explicit rate limits, timeouts, and resource quotas for API calls, file operations, and compute. Implement circuit breakers for runaway loops.
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 | $0.00100 | $0.01865 |
| Opus 5.5 | $0.00040 | $0.00746 |
| Sonnet 5.5 | $0.00020 | $0.00373 |
| Haiku 4.5 | $0.00010 | $0.00186 |
Grade A, and why
rbac-audit 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Zava Learning — RBAC / Least-Privilege & Access-Usage Audit
Resource Group: @@RG@@. Read-only audit — never create, modify, or remove a role assignment.
This is a weekly governance review of who can do what here, and whether they actually use it. The
goal is real least-privilege hygiene: surface over-privileged identities, direct (ungoverned)
assignments, and standing access that has gone unused so a human can clean it up.
Identity naming policy (governance reports name their subjects)
The principals under audit (their display name / UPN / objectId and role) are the subject of this
report — show them; an RBAC finding that hides the identity is useless. This is NOT the learner-PII
the redaction standard protects. Still apply SearchMemory("zava-redaction") redact() to strip any
secrets, tokens, credentials, or learner/customer PII that happen to appear, and never print
secret values — but DO name the admins, service principals, and groups being audited.
Where the names come from (do NOT call Microsoft Graph): take the identity name from the
principalName field that az role assignment list already returns, plus principalType. Do not
run az ad user/group/sp list|show or az rest https://graph.microsoft.com/... to resolve names or
look up guest/enabled status — the SRE Agent's managed identity has no Graph directory-read
permission, so those calls fail with Insufficient privileges AND are gated as PendingAuthorization,
which would stall this unattended weekly run. If a principalName is blank, identify that principal by
its principalId (objectId) + principalType and note "display name unavailable (directory lookup not
permitted)". A UPN containing #EXT# (when present in principalName) still indicates a guest.
1. Enumerate the real assignments effective on the RG
- All assignments scoped at or above the RG (this includes subscription/management-group inheritance —
most access here is inherited):
az role assignment list --scope /subscriptions/<sub>/resourceGroups/@@RG@@ --include-inherited -o json. - For each: capture
principalName,principalType(User / Group / ServicePrincipal),roleDefinitionName,scope,description, andcreatedOnwhere present. These fields are sufficient — everything you need to name and classify a principal is already in this output; no follow-up directory calls.
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.
- 1mo ago First seen · 103 lines · 100 tokens per session scan A a31640482a8e
rbac-audit is a skill published in the GitHub repository microsoft/sre-agent (179 stars, last pushed 3d ago), licensed MIT. It adds 100 tokens to every session and 1,865 once invoked, about $0.0004 per session on Opus 5.5. 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 skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…