MCPInstall

neo4j-driver-dotnet-skill

Neo4j .NET Driver v6 — IDriver lifecycle, DI registration (singleton), ExecutableQuery fluent API, ExecuteReadAsync/ExecuteWriteAsync managed transactions, IResultCursor (FetchAsync/ ToListAsync), record value access (.Get /As ), null safety, UNWIND batching, temporal types, await using, EagerResult, object mapping, CancellationToken, error handling, and common traps. Use when writing C or .NET code connecting to Neo4j. Also triggers on Neo4j.Driver, IDriver, ExecutableQuery, ExecuteReadAsync, ExecuteWriteAsync, IResultCursor, IAsyncSession, or any Bolt/Aura work in .NET/C. Does NOT handle Cypher authoring — use neo4j-cypher-skill. Does NOT cover driver version upgrades — use neo4j-migration-skill.

Installs
767
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/neo4j-contrib/neo4j-skills --skill neo4j-driver-dotnet-skill
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads neo4j-driver-dotnet-skill automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

When to Use

  • Writing C# or .NET code connecting to Neo4j
  • Setting up IDriver, DI registration, or session/transaction lifecycle
  • Questions about ExecutableQuery, IResultCursor, async patterns, result mapping
  • Debugging sessions, type mapping, null safety, or error handling in .NET

When NOT to Use

  • Writing/optimizing Cypher queries → neo4j-cypher-skill
  • Upgrading from older driver version → neo4j-migration-skill

Install

dotnet add package Neo4j.Driver
PackageUse
Neo4j.DriverAsync API — use this
Neo4j.Driver.SimpleSynchronous wrapper
Neo4j.Driver.ReactiveSystem.Reactive streams

Driver Lifecycle

IDriver — thread-safe, connection-pooled, expensive to create. Create one per application.

using Neo4j.Driver;

// URI schemes:
//   neo4j+s://xxx.databases.neo4j.io   — TLS + cluster routing (Aura)
//   neo4j://localhost                   — unencrypted + cluster routing
//   bolt+s://localhost:7687             — TLS + single instance
//   bolt://localhost:7687               — unencrypted + single instance

await using var driver = GraphDatabase.Driver(
    "neo4j+s://xxx.databases.neo4j.io",
    AuthTokens.Basic("neo4j", "password"));

await driver.VerifyConnectivityAsync();   // fail fast on startup

IDriver and IAsyncSession implement IAsyncDisposable — always await using, never plain using.

// ❌ Wrong — synchronous Dispose() may block thread pool
using var driver = GraphDatabase.Driver(uri, auth);

// ✅ Correct
await using var driver = GraphDatabase.Driver(uri, auth);

Auth options: AuthTokens.Basic(u, p) / AuthTokens.Bearer(token) / AuthTokens.Kerberos(ticket) / AuthTokens.None


Environment Variables

Load connection config from environment / appsettings.json — never hardcode credentials.

// appsettings.json
{
  "Neo4j": {
    "Uri": "neo4j+s://xxx.databases.neo4j.io",
    "User": "neo4j",
    "Password": "secret",
    "Database": "neo4j"
  }
}
// Access via IConfiguration (injected in Program.cs)
var uri      = builder.Configuration["Neo4j:Uri"];
var user     = builder.Configuration["Neo4j:User"];
var password = builder.Configuration["Neo4j:Password"];
var database = builder.Configuration["Neo4j:Database"] ?? "neo4j";

Override with environment variables (standard .NET behavior): Neo4j__Uri=neo4j+s://... (double underscore = colon separator). Never commit appsettings.json with real credentials — use appsettings.Development.json (gitignored) or env vars in CI/production.


DI Registration (ASP.NET Core)

Register IDriver as singleton — never Scoped or Transient. Never register IAsyncSession in DI.

// Program.cs
builder.Services.AddSingleton<IDriver>(_ =>
    GraphDatabase.Driver(
        builder.Configuration["Neo4j:Uri"],
        AuthTokens.Basic(
            builder.Configuration["Neo4j:User"],
            builder.Configuration["Neo4j:Password"])));

// Shutdown hook — dispose the singleton cleanly
builder.Services.AddHostedService<Neo4jShutdownService>();

// Neo4jShutdownService.cs
public class Neo4jShutdownService(IDriver driver, IHostApplicationLifetime lifetime)
    : IHostedService
{
    public Task StartAsync(CancellationToken _)
    {
        lifetime.ApplicationStopping.Register(() =>
            driver.DisposeAsync().AsTask().GetAwaiter().GetResult());
        return Task.CompletedTask;
    }
    public Task StopAsync(CancellationToken _) => Task.CompletedTask;
}

// Inject into services — sessions opened per unit of work
public class PersonService(IDriver driver)
{
    public async Task<List<string>> GetNamesAsync(CancellationToken ct = default)
    {
        var (records, _, _) = await driver
            .ExecutableQuery("MATCH (p:Person) RETURN p.name AS name")
            .WithConfig(new QueryConfig(database: "neo4j"))
            .ExecuteAsync(ct);
        return records.Select(r => r.Get<string>("name")).ToList();
    }
}

Choose the Right API

APIWhenAuto-retryStreaming
driver.ExecutableQuery()Most queries — simple default✅❌ eager
session.ExecuteReadAsync/WriteAsync()Large results, multi-query tx✅✅
session.RunAsync()LOAD CSV, CALL {} IN TRANSACTIONS❌✅
session.BeginTransactionAsync()Multi-function, external coordination❌✅

ExecutableQuery — Recommended Default

Fluent builder; manages session, transaction, retries, and bookmarks automatically.

// Read
var (records, summary, keys) = await driver
    .ExecutableQuery("MATCH (p:Person {name: $name})-[:KNOWS]->(f) RETURN f.name AS name")
    .WithParameters(new { name = "Alice" })
    .WithConfig(new QueryConfig(
        database: "neo4j",
        routing: RoutingControl.Readers))    // route reads to replicas
    .ExecuteAsync(cancellationToken);

foreach (var r in records)
    Console.WriteLine(r.Get<string>("name"));

// Use ResultConsumedAfter for wall-clock timing (ResultAvailableAfter = time-to-first-byte only)
Console.WriteLine($"{summary.ResultConsumedAfter.TotalMilliseconds} ms");

// Write
var (_, writeSummary, _) = await driver
    .ExecutableQuery("CREATE (p:Person {name: $name, age: $age})")
    .WithParameters(new { name = "Bob", age = 30 })
    .WithConfig(new QueryConfig(database: "neo4j"))
    .ExecuteAsync();
Console.WriteLine($"Created {writeSummary.Counters.NodesCreated} nodes");

// WithMap — project inline
var names = await driver
    .ExecutableQuery("MATCH (p:Person) RETURN p.name AS name")
    .WithConfig(new QueryConfig(database: "neo4j"))
    .WithMap(r => r["name"].As<string>())
    .ExecuteAsync();   // names.Result is IReadOnlyList<string>

Never await omitted: ExecuteAsync() returns Task — missing await compiles silently but query never runs.

Never string-interpolate Cypher. Always WithParameters() — prevents injection, enables plan caching.


Managed Transactions

Use for large result sets (lazy streaming) or multiple queries per transaction. Callback auto-retried on transient failure — keep it idempotent, no side effects inside.

await using var session = driver.AsyncSession(conf => conf.WithDatabase("neo4j"));

// Read — routes to replicas
var names = await session.ExecuteReadAsync(async tx =>
{
    var cursor = await tx.RunAsync(
        "MATCH (p:Person) WHERE p.name STARTS WITH $prefix RETURN p.name AS name",
        new { prefix = "Al" });
    return await cursor.ToListAsync(r => r.Get<string>("name"));
    // Consume cursor INSIDE callback — invalid after callback returns
});

// Write — void, no async needed
await session.ExecuteWriteAsync(tx =>
    tx.RunAsync("MERGE (p:Person {name: $name})", new { name = "Carol" }));

// Write — async when needing counters
var summary = await session.ExecuteWriteAsync(async tx =>
{
    var cursor = await tx.RunAsync(
        "CREATE (p:Person {name: $name})", new { name = "Alice" });
    return await cursor.ConsumeAsync();   // drains cursor, returns IResultSummary
});
Console.WriteLine($"Created {summary.Counters.NodesCreated} nodes");

Cursor rules:

  • Consume with ToListAsync() or FetchAsync() loop inside the callback
  • Returning a cursor from the callback → transaction closes → cursor invalid → exception
// ❌ Returns cursor — tx closes immediately after lambda returns
var cursor = await session.ExecuteReadAsync(async tx =>
    await tx.RunAsync("MATCH (p:Person) RETURN p.name AS name"));
await cursor.FetchAsync();   // throws

// ✅ Consume inside
var names = await session.ExecuteReadAsync(async tx =>
{
    var cursor = await tx.RunAsync("MATCH (p:Person) RETURN p.name AS name");
    return await cursor.ToListAsync(r => r.Get<string>("name"));
});

Async void trap:

// ❌ CS1998 warning — async with no await; RunAsync Task discarded
await session.ExecuteWriteAsync(async tx =>
    tx.RunAsync("MERGE (p:Person {name: $name})", new { name = "Alice" }));

// ✅ No async, return Task directly
await session.ExecuteWriteAsync(tx =>
    tx.RunAsync("MERGE (p:Person {name: $name})", new { name = "Alice" }));

FetchAsync Loop

var cursor = await tx.RunAsync("MATCH (p:Person) RETURN p.name AS name");

while (await cursor.FetchAsync())          // true while records remain
{
    Process(cursor.Current.Get<string>("name"));
}
// Do NOT use cursor.Current after the loop — it holds the last record, not null
// Do NOT call FetchAsync() again after it returned false — throws InvalidOperationException

Cursor consumption methods:

MethodRecordsSummaryUse
ToListAsync()✅ all❌Need records
ToListAsync(mapper)✅ mapped❌Need mapped records
FetchAsync() loop✅ one/time❌ until ConsumeAsyncLarge/lazy
ConsumeAsync()❌ discards✅Need counters

IResultSummary profiling [v6.3.0+]: summary.Profile is [Obsolete] → use summary.QueryProfile (IQueryProfile), whose DbHits, Rows, Time, PageCacheHits, PageCacheMisses are long? — null means not recorded, distinct from 0. Gate on summary.HasProfile. | SingleAsync() | ✅ exactly 1 | ❌ | Expect one row |


Record Value Access

// Two equivalent patterns — prefer .Get<T>()
string name = record.Get<string>("name");
int    age  = record.Get<int>("age");

string name2 = record["name"].As<string>();   // indexer + As<T>
string name3 = record[0].As<string>();        // by column index

// Null safety — .As<T>() on null graph value throws InvalidCastException
string? city = record["city"].As<string?>();  // ✅ nullable
int?    age2 = record["age"].As<int?>();      // ✅ nullable

// Absent key — throws KeyNotFoundException (typo or not in RETURN)
if (record.Keys.Contains("city"))
    var city3 = record.Get<string?>("city");

Type Mapping

Cypher.NET defaultNotes
Integerlongsafe: int, long?, int?
Floatdoublesafe: float, double?
Stringstringuse string? if nullable
Booleanbool
ListIReadOnlyList<object>
MapIReadOnlyDictionary<string,object>
NodeINode.Labels, .Properties, .ElementId
RelationshipIRelationship.Type, .StartNodeElementId
DateLocalDate.ToDateOnly() (.NET 6+)
DateTimeZonedDateTime.ToDateTimeOffset() (ms precision)
LocalDateTimeLocalDateTime
DurationDuration.ToTimeSpan() throws if has months/days
UUIDSystem.Guid[driver 6.2+, Bolt 6.1, Neo4j 2026.08+]; object mapping also converts to/from string
nullnulluse nullable types

ElementId stable within one transaction only — do not use to MATCH across separate transactions.

// Pass CLR types as params — driver converts automatically
await driver.ExecutableQuery("CREATE (e:Event {at: $ts})")
    .WithParameters(new { ts = DateTimeOffset.UtcNow })
    .WithConfig(new QueryConfig(database: "neo4j"))
    .ExecuteAsync();

UNWIND Batching

// ❌ One transaction per record — high overhead
foreach (var item in items)
    await driver.ExecutableQuery("MERGE (n:Node {id: $id})")
        .WithParameters(new { id = item.Id })
        .WithConfig(new QueryConfig(database: "neo4j"))
        .ExecuteAsync();

// ✅ Single transaction via UNWIND — anonymous types only (custom classes don't serialize)
var rows = items.Select(i => new { id = i.Id, name = i.Name }).ToArray();
await driver.ExecutableQuery(@"
    UNWIND $rows AS row
    MERGE (n:Node {id: row.id})
    SET n.name = row.name")
    .WithParameters(new { rows })
    .WithConfig(new QueryConfig(database: "neo4j"))
    .ExecuteAsync();

Custom class instances passed to WithParameters for UNWIND do not serialize — use new object[] { new { ... } } or Dictionary<string, object>.


Object Mapping (Preview API)

using Neo4j.Driver.Preview.Mapping;   // REQUIRED — without this, AsObject<T>() is CS1061

public record Person(string Name, int Age);   // C# records work well here

var result = await driver
    .ExecutableQuery("MATCH (p:Person) RETURN p.name AS name, p.age AS age")
    .WithConfig(new QueryConfig(database: "neo4j"))
    .ExecuteAsync();

var person = result.Result[0].AsObject<Person>();   // RETURN keys map to record properties

// Bulk mapping
var (people, _, _) = await driver
    .ExecutableQuery("MATCH (p:Person) RETURN p.name AS name, p.age AS age")
    .WithConfig(new QueryConfig(database: "neo4j"))
    .AsObjectsAsync<Person>();

Error Handling

try
{
    await driver.ExecutableQuery("...")
        .WithConfig(new QueryConfig(database: "neo4j"))
        .ExecuteAsync();
}
catch (AuthenticationException ex)      { /* bad credentials */ }
catch (ServiceUnavailableException ex)  { /* database unreachable */ }
catch (ClientException ex)
    when (ex.Code == "Neo.ClientError.Schema.ConstraintValidationFailed")
{
    // Unique/existence constraint violation — catch BEFORE Neo4jException
}
catch (Neo4jException ex)               { /* all other server errors */ }

Catch ClientException before Neo4jException — it's a subclass; generic handler swallows it.

ex.GqlStatus — stable GQL status codes; prefer over string-matching ex.Code.

Explicit transaction rollback can itself throw — isolate it:

catch (Exception original)
{
    try { await tx.RollbackAsync(); }
    catch (Exception ex) { logger.LogError(ex, "Rollback failed"); }
    throw;
}

If CommitAsync() throws a network error, commit may or may not have succeeded — design writes idempotent with MERGE + unique constraints.


Common Mistakes

MistakeFix
using var driverawait using var driver — IDriver is IAsyncDisposable
using var sessionawait using var session
IDriver as Scoped/Transient in DIRegister as Singleton
IAsyncSession in DINever — open per unit of work
Missing await on ExecuteAsync()Task silently never runs
async tx => tx.RunAsync(...) no inner awaitRemove async, return Task directly
Omit database in QueryConfig/AsyncSessionAlways specify — saves a round-trip
No CancellationToken in web appsPropagate HttpContext.RequestAborted
.As<string>() on null graph value.As<string?>() — non-nullable throws
record["key"] absent keyCheck record.Keys.Contains() first
cursor.Current after FetchAsync loopLast record, not null — don't use after loop
FetchAsync() after false returnThrows — stop loop, don't call again
Return cursor from managed tx callbackConsume with ToListAsync() inside callback
Need counters from session writeawait cursor.ConsumeAsync()
AsObject<T>() CS1061 compile errorAdd using Neo4j.Driver.Preview.Mapping;
ResultAvailableAfter for total timingUse ResultConsumedAfter (full wall-clock)
Custom class in WithParameters for UNWINDUse anonymous types or Dictionary<string,object>
Rename C# param but not Cypher $paramAnonymous property names must match $param names
ExecuteWriteAsync for readsUse ExecuteReadAsync — routes to replicas
Side effects inside managed tx callbackMove outside — callback retried on failure
Duration.ToTimeSpan() with months/daysOnly safe for pure second/nanosecond durations
Catch Neo4jException before ClientExceptionClientException is subclass — catch it first

References

Load on demand:

  • references/transactions.md — explicit transactions, BeginTransactionAsync, rollback, commit uncertainty, TransactionConfig (timeout, metadata), causal consistency and bookmarks
  • references/performance.md — spatial types (Point/WGS-84/Cartesian), connection pool tuning, WithFetchSize, session config options, CancellationToken patterns, large result streaming
  • references/object-mapping.md — AsObject<T>, blueprint mapping, lambda mapping, AsObjectsAsync<T>, repository pattern example

Checklist

  • IDriver registered as singleton in DI (or await using for short-lived apps)
  • await using on driver and sessions (not plain using)
  • database specified in QueryConfig / AsyncSession config
  • ExecutableQuery used for simple queries; ExecuteReadAsync/ExecuteWriteAsync for streaming/multi-query
  • Cursor consumed inside managed tx callback (not returned)
  • Nullable types (string?, int?) on any graph value that can be null
  • WithParameters() used (no string interpolation)
  • UNWIND batching with anonymous types (not custom class instances)
  • CancellationToken propagated in web app handlers
  • ClientException caught before Neo4jException
  • Writes idempotent (MERGE + constraints) for retry safety
  • No side effects inside ExecuteReadAsync/ExecuteWriteAsync callbacks

FAQ

What does the neo4j-driver-dotnet-skill skill do?

Neo4j .NET Driver v6 — IDriver lifecycle, DI registration (singleton), ExecutableQuery fluent API, ExecuteReadAsync/ExecuteWriteAsync managed transactions, IResultCursor (FetchAsync/ ToListAsync), record value access (.Get /As ), null safety, UNWIND batching, temporal types, await using, EagerResult, object mapping, CancellationToken, error handling, and common traps. Use when writing C or .NET code connecting to Neo4j. Also triggers on Neo4j.Driver, IDriver, ExecutableQuery, ExecuteReadAsync, ExecuteWriteAsync, IResultCursor, IAsyncSession, or any Bolt/Aura work in .NET/C. Does NOT handle Cypher authoring — use neo4j-cypher-skill. Does NOT cover driver version upgrades — use neo4j-migration-skill.

How do I install neo4j-driver-dotnet-skill?

Run npx skills add https://github.com/neo4j-contrib/neo4j-skills --skill neo4j-driver-dotnet-skill in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/neo4j-driver-dotnet-skill/ (personal) or .claude/skills/neo4j-driver-dotnet-skill/ (project).

Which AI agents can use neo4j-driver-dotnet-skill?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is neo4j-driver-dotnet-skill safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains neo4j-driver-dotnet-skill?

It is published by neo4j-contrib in neo4j-contrib/neo4j-skills. It has 767 recorded installs.

More from neo4j-contrib/neo4j-skills

View all →

neo4j-cypher-skill

neo4j-contrib/neo4j-skills

Generates, optimizes, and validates Cypher 25 queries for Neo4j 2025.x and 2026.x. Use when writing new Cypher queries, optimizing slow queries, graph pattern matching, vector or fulltext search, subqueries, or batch writes. Covers MATCH, MERGE, CREATE, WITH, RETURN, CALL, UNWIND, FOREACH, LOAD CSV, SEARCH, expressions, functions, indexes, and subqueries. Does NOT handle driver migration or API changes — use neo4j-migration-skill. Does NOT cover DB administration or server ops — use neo4j-cli-tools-skill.

Backend & APIs1.5K installs

neo4j-modeling-skill

neo4j-contrib/neo4j-skills

Design, review, and refactor Neo4j graph data models. Use when choosing node labels vs relationship types vs properties, migrating relational/document schemas to graph, detecting anti-patterns (generic labels, supernodes, missing constraints), designing intermediate nodes for n-ary relationships, enforcing schema with constraints and indexes, or assessing an existing model against graph modeling best practices. Does NOT handle Cypher query authoring — use neo4j-cypher-skill. Does NOT handle Spring Data Neo4j entity mapping — use neo4j-spring-data-skill. Does NOT handle GraphQL type definitions — use neo4j-graphql-skill. Does NOT handle data import — use neo4j-import-skill.

Backend & APIs1.2K installs

neo4j-graphrag-skill

neo4j-contrib/neo4j-skills

Build GraphRAG retrieval pipelines on Neo4j using the neo4j-graphrag Python package (v1.22.0+). Covers retriever selection (VectorRetriever, HybridRetriever, VectorCypherRetriever, HybridCypherRetriever, Text2CypherRetriever, ToolsRetriever), external vector DB retrievers (Weaviate, Pinecone, Qdrant), retrievalquery Cypher fragments, queryparams, filters, GraphRAG pipeline wiring (GraphRAG + LLM + prompt), all LLM providers (OpenAI, Anthropic, Gemini/VertexAI, Bedrock, Cohere, Mistral, Ollama), embedder setup, index creation, token usage tracking, Cypher 25 SEARCH clause, and LangChain/LlamaIndex integration. Does NOT handle KG construction — use neo4j-document-import-skill. Does NOT handle plain vector search — use neo4j-vector-index-skill. Does NOT handle GDS analytics — use neo4j-gds-skill. Does NOT handle agent memory — use neo4j-agent-memory-skill.

AI & Agents1.1K installs

neo4j-driver-python-skill

neo4j-contrib/neo4j-skills

Neo4j Python Driver v6 — driver lifecycle, executequery, managed and explicit transactions, async (AsyncGraphDatabase), result handling, data type mapping, error handling, UNWIND batching, connection pool tuning, and causal consistency. Use when writing Python code that connects to Neo4j via GraphDatabase.driver, executequery, executeread, executewrite, AsyncGraphDatabase, neo4j.Result, or RoutingControl. Package name is neo4j (not neo4j-driver) since v6. Python =3.10 required. Does NOT handle Cypher query authoring — use neo4j-cypher-skill. Does NOT cover driver upgrades or breaking changes — use neo4j-migration-skill. Does NOT cover GraphRAG pipelines (neo4j-graphrag package) — use neo4j-graphrag-skill.

neo4j-document-import-skill

neo4j-contrib/neo4j-skills

Ingests unstructured and semi-structured documents into Neo4j as a knowledge graph. Use when chunking PDFs, HTML, plain text, or Markdown; extracting entities and relationships from text with an LLM (SimpleKGPipeline, neo4j-graphrag); loading JSON via apoc.load.json; building Document→Chunk→Entity graph structures; or connecting LangChain/LlamaIndex document loaders to Neo4j. Covers neo4j-graphrag SimpleKGPipeline, LLM Graph Builder web UI, entity resolution, chunking strategies, and graph schema design for RAG pipelines. Does NOT handle structured CSV/relational import — use neo4j-import-skill. Does NOT handle GraphRAG retrieval after ingestion — use neo4j-graphrag-skill. Does NOT handle vector index creation — use neo4j-vector-search-skill.

AI & Agents934 installs

neo4j-query-tuning-skill

neo4j-contrib/neo4j-skills

Diagnoses and fixes slow Neo4j Cypher queries by reading execution plans, identifying bad operators (AllNodesScan, CartesianProduct, Eager, NodeByLabelScan), and prescribing fixes (indexes, hints, query rewrites, runtime selection). Use when a query is slow, when EXPLAIN or PROFILE output needs interpretation, when dbHits or pageCacheHitRatio are poor, when cardinality estimation diverges from actuals, or when deciding between slotted/pipelined/parallel runtimes. Covers USING INDEX / USING SCAN / USING JOIN hints, db.stats.retrieve, SHOW QUERIES, SHOW TRANSACTIONS, TERMINATE TRANSACTION. Does NOT write new Cypher from scratch — use neo4j-cypher-skill. Does NOT cover GDS algorithm tuning — use neo4j-gds-skill. Does NOT cover index/constraint creation syntax details — use neo4j-cypher-skill references/indexes.md.

916 installs

Popular backend & apis skills

View category →

find-skills

vercel-labs/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

OfficialBackend & APIs4.1M installs

entra-app-registration

microsoft/azure-skills

Guides Microsoft Entra ID app registration, OAuth 2.0 authentication, and MSAL integration. USE FOR: create app registration, register Azure AD app, configure OAuth, set up authentication, add API permissions, generate service principal, MSAL example, console app auth, Entra ID setup, Azure AD authentication. DO NOT USE FOR: Key Vault secrets (use azure-keyvault-expiration-audit), general Azure resource security guidance.

OfficialBackend & APIs566.7K installs

This skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use state management" — in a project that needs ADK (Agent Development Kit) API patterns and code examples. It provides a quick reference for agent types, tool definitions, orchestration patterns, callbacks, state management, the graph Workflow API, and reference recipes to study. Do NOT use for scaffolding (use google-agents-cli-scaffold) or deployment (use google-agents-cli-deploy).

Backend & APIs514.8K installs

embedded-captions

heygen-com/hyperframes

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet anchor rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.

Backend & APIs453.3K installs

entra-agent-id

microsoft/azure-skills

Provision Microsoft Entra Agent Identity Blueprints, BlueprintPrincipals, and per-instance Agent Identities via Microsoft Graph, and configure OAuth 2.0 token exchange (fmipath, OBO, cross-tenant) including the Microsoft Entra SDK for AgentID sidecar. USE FOR: Agent Identity Blueprint, BlueprintPrincipal, agent OAuth, fmipath token exchange, agent OBO, Workload Identity Federation for agents, polyglot agent auth, Microsoft.Identity.Web.AgentIdentities. DO NOT USE FOR: standard Entra app registration (use entra-app-registration), Microsoft Foundry agent authoring (use microsoft-foundry).

OfficialBackend & APIs336.6K installs

neon

neondatabase/agent-skills

Overview of Neon, a complete set of cloud backend primitives around Lakebase Postgres: Auth, Object Storage, Functions, the AI Gateway, and Realtime. Start here to choose Neon for undecided login, files, APIs, and LLM calls, set up the CLI or MCP server, and follow the branch-first workflow. Use when building an app or backend on Neon, or when "Neon" or "Lakebase Postgres" is mentioned. Child skill neon-postgres wins for an existing DATABASEURL, SQL, schema, inspect, or search. Child skill neon-auth wins for login, users, sessions, identity routing, and Managed Better Auth setup. Child skill neon-realtime wins for live queries in reactive apps and agents. Also use for object storage, S3, buckets, serverless functions, function triggers, cron, AI gateway, LLM calls, realtime, logs, Loki, Grafana, observability, postgres, database, backend, Claimable Neon, neon.new, or a no-signup database.

Backend & APIs237.1K installs