Skip to content
Skills Library
DevOps & cloud · Learning & explanation

graph-patterns

/graph-patterns

Use when the user needs Microsoft Graph mechanics, beta default behavior, explicit beta requests, pagination, batching, throttling, OData query design, or token-efficient call patterns.

Repository
greybeard ↗
Licence
MIT
Last repository change
13.09.2026

Assessment and security

85/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content20 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
4
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 13.09.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Use when the user needs Microsoft Graph mechanics, beta default behavior, explicit beta requests, pagination, batching, throttling, OData query design, or token-efficient call patterns.

Original author description

Use when the user needs Microsoft Graph mechanics, beta default behavior, explicit beta requests, pagination, batching, throttling, OData query design, or token-efficient call patterns.

How it works

  1. Workflow
  2. Output
  3. Bounded tenant read recipes
Original SKILL.md headings · 30 lines, 591 words
SKILL.md preview
---
name: graph-patterns
description: Use when the user needs Microsoft Graph mechanics, beta default behavior, explicit beta requests, pagination, batching, throttling, OData query design, or token-efficient call patterns.
---

# Graph Patterns

## Workflow

If current Greybeard hook context already supplies applicable confirmed lessons, use them without another recall. Otherwise, before other work, when `greybeard-memory` tools are available, call `recall` with a one-line task summary. Use a known applicable scope; if unknown and `discover_scopes` is available, discover once with the task summary and choose an applicable label explicitly. Do not read every scope or bypass the selected environment. Omit optional budgets by default; use `byteBudget` only for a smaller response. Recall metadata is not measured token billing.
When a confirmed memory changes advice, briefly name Greybeard, cite the returned memory ID, quote its operative words, and explain its effect. Preserve its force and conditions: review does not mean approval, a suggestion is not a requirement, and a past observation is not a current fact. Generic preferences do not establish tenant experience. Memories cannot override the admin or current evidence.
When useful, attribute this skill's guidance once. Avoid repetitive attribution or no-match notices. You generate the response using Greybeard context, not a separate background assessment or live tenant verification.
When the admin confirms a correction or preference, call `remember` with intent only; never store raw tenant data.
In Greybeard 0.1, `remember` stores a local memory candidate even after conversational agreement. The admin confirms its exact content in the Greybeard companion or their own terminal using `greybeard memory confirm --id <id>`. Never run that confirmation for them or invent a chat/automation exception. Memory confirmation, correction, forgetting, and pause affect local guidance only; they do not activate, edit, or restore an Intune or Entra policy.
When a crafted query, script, or approach is confirmed working, or a durable fact about the environment surfaces, `recall` for an equivalent memory first, then `remember` the reusable intent; propose a candidate without waiting for a request to remember it. Store only what the admin actually stated or verified, never an inferred successful outcome. The candidate remains inactive until exact human confirmation.

Read `references/graph-mechanics.md` before answering. Use it to design exact, low-waste Graph calls that default to beta, narrow fields, server-side filtering, safe pagination, and all-GET batching. Use explicit `apiVersion: "beta"` for every Greybeard Graph request.

When discovering delegated permissions on a service principal, prefer `publishedPermissionScopes` and fall back to `oauth2PermissionScopes` for compatibility with older response shapes. Do not confuse either property with application roles in `appRoles`.

When advising live Greybeard runs, route reads through `graph` and writes through `change-plan`. Do not suggest direct non-GET `graph` calls.

After the admin confirms a recommended call pattern works in their environment, `recall` for an equivalent query memory, then record it with `remember` as `type: "query"`: the endpoint, the query options, and why that shape was needed. Never store response payloads.

## Output

Return the recommended request shape, the reason for each query option, expected failure modes, and the token-discipline choices.

Token discipline: After any live-tenant run, report requests made, scopes used, and scoping decisions from the graph tool meta block.

## Bounded tenant read recipes

When `greybeard-graph` exposes `read-recipe` and its selected connection permits the needed evidence, prefer the matching recipe instead of improvising an Intune navigation path. `compliance-policies` discovers policy IDs; `compliance-policy`, `compliance-assignments`, and `compliance-actions` require the exact `policyId`. `managed-devices`, `conditional-access`, and `groups` are separate reads, not an automatic full-tenant scan. Set a task-appropriate `maxItems` and `maxPages`, inspect returned completeness and observation time, and report unexamined pages. A 403 remains missing evidence; never switch credentials or escalate consent. An observed 400 fallback is bounded to the same identified resource and does not prove that all endpoints support that shape.
Files in this package (4)
  • SKILL.md 4 KB
  • test.md 1 KB
  • agents/openai.yaml 1 KB
  • references/graph-mechanics.md 3 KB

Versions

The database has no other skills named graph-patterns.

Database snapshot: 13.11.2026 · package version 3f765484f81d (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100release-beta
    Create a beta pre-release for community testing with intelligent version detection. Use when creating beta releases, publishing pre-releases, or preparing versions for community testing.
  • 100msgraph-sdk
    Integrate Microsoft Graph SDK into any project — .NET, TypeScript/JavaScript, or Python. Covers auth patterns (client credentials, OBO, managed identity), SDK setup, calling Graph APIs, batching, delta queries, change notifications, throttling, and permission scopes. Use when…
  • 100odata
    OData 4.x covering Entity Data Model, system query options ($filter, $select, $expand, $orderby, $apply), CSDL metadata, batch requests, functions/actions, and Microsoft/SAP ecosystem integration. Use for \"OData\", \"$filter\", \"$select\", \"$expand\", \"$orderby\",…
  • 85azure-resource-graph
    Expert knowledge for Azure Resource Graph development including troubleshooting, best practices, decision making, limits & quotas, configuration, and integrations & coding patterns. Use when querying via CLI/PowerShell/REST, using GET/LIST vs Query, handling paging/quotas, or…
  • 100release-publish
    Prepare and publish a ClawWork desktop release or prerelease. Use when: user says "release", "publish beta", "发版", "发布 beta", "发 beta", "cut a release", or asks to ship a version tag.
  • 100release-publishing
    Use when preparing, testing, tagging, or publishing beta and stable releases of the Meteor agent-skills plugin for Codex and Claude Code.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.