SkillAgentSearch skills...

outsystems-mcp

Plugins for the OutSystems remote MCP.

Install / Use

npx skills add OutSystems/outsystems-mcp

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

57/100

Supported Platforms

Universal

Our assessment of outsystems-mcp

outsystems-mcp scores 57/100 on our quality scale, 4227th of 4,698 Development & Engineering skills we index.

Its SKILL.md is 57 KB long, well organised into 11 sections and no code examples: long enough that it reads more like full documentation than a focused instruction file, which agents can find harder to follow.

It has no GitHub stars yet, so there is no community track record; judge it on its content.

Substance
21/30
Structure
13/20
Description
8/15
Adoption
0/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated 3 days ago, so outsystems-mcp is actively maintained.
  • No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
  • Its trust signals score 80/100, with 2 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-11-16. Automated pattern scan on 2026-11-16. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

outsystems-mcp compared with similar skills

All 4 of these similar skills score higher than outsystems-mcp; compare them before choosing.

SkillScoreStarsUpdatedFormat
outsystems-mcp (this skill)by OutSystems5703d agoSKILL.md
taste-skillby Leonxlnx100116.8k8d agoCLAUDE.md
Agent-Reachby Panniantong100112.3k8d agoCLAUDE.md
rufloby ruvnet10081.9ktodayCLAUDE.md
headroomby headroomlabs-ai10079.8ktodayCLAUDE.md

Frequently asked questions

How do I install outsystems-mcp?
Run npx skills add OutSystems/outsystems-mcp. The install tabs above show the steps for each supported agent.
Which AI agents does outsystems-mcp work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is outsystems-mcp safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 80/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is outsystems-mcp still maintained?
The repository was last updated 3 days ago, so outsystems-mcp is actively maintained.

OutSystems - Remote MCP

You are connected to OutSystems Developer Cloud (ODC) over the MCP HTTP transport. OutSystems is a cloud-native low-code platform where apps are built from OML (OutSystems Model Language), a binary format describing entities, screens, actions, and logic. Every tool call carries the harness's validated OAuth bearer; tenant + user identity are derived from the JWT, not from arguments.

Once authenticated and the server's tools are visible, read the live tools/list before your first non-auth OutSystems operation. This skill names domains, not tools — names, parameters, and defaults can change server-side, and the server is the source of truth. (Authenticating is the one exception: drive the auth flow first, as described below.)

Platform check, only when both OutSystems MCP servers are connected. This server is for OutSystems Developer Cloud (ODC). When tools from the Service Studio MCP server (O11, named servicestudio in the O11 setup) are also visible in this session, you MUST decide the platform before your first call to this server, read-only calls such as listing apps included. With both servers connected, you can't know which platform a request is for unless it names one, so never guess:

  • The request names ODC, an ODC app or the tenant: go ahead on ODC.
  • The request names O11, an O11 module or the module open in Service Studio: don't use this server.
  • The request names neither, as in "my app": you MUST ask the user "Is this for ODC, or for the O11 module open in Service Studio?" and wait for the answer before calling any tool.

Once you know the platform for certain, because the request named it, the user answered, or earlier turns in this conversation settled it, don't ask again for later requests; ask again only when a request points to the other platform.

When only this server is connected, skip this check and go ahead as usual.

Authenticating

The remote MCP server is OAuth-protected with standard OAuth: an unauthenticated call gets 401 + WWW-Authenticate, and the server advertises OAuth discovery + dynamic client registration (/authorize, /token, /register, PKCE S256). Authentication is performed by your MCP client, not by an OutSystems tool — the server exposes no authenticate tool. How you sign in depends on the harness:

  • GitHub Copilot / VS Code (and most other MCP clients) expose no agent-callable auth tool — the IDE runs the OAuth flow in its own UI (opens the browser, captures the callback). Just make the first OutSystems tool call; the client prompts the user to sign in. Tell the user to complete that prompt, then proceed. Don't look for an authenticate tool — it isn't there.
  • Kiro drives the sign-in through its own MCP UI — same as the IDE clients above, no authenticate tool.
  • CLI-based clients (Cursor CLI, Copilot CLI) open the system browser and listen on localhost for the callback, allowing shell-based recovery if needed.

If the browser shows "site can't be reached" at the callback URL: The sign-in succeeded on the tenant side — only the callback connection failed, typically an IPv4/IPv6 loopback mismatch in a VM. Do not make the user wait for timeout; tell them immediately. Ask for the full callback URL from the address bar. Recovery depends on your harness: some clients can replay the callback via shell commands; others require the user to adjust the address in their browser. See your harness's specific Authenticating section for the exact steps. Handle authorization codes as sensitive values — code is single-use and short-lived.

Lazy. Authenticate before the first OutSystems tool call, following your harness's path above. The real tools appear once the user has authorized — wait for that, then proceed. Your harness completes the sign-in in its own UI; you won't receive an authorization URL to relay.

Reactive. On data.category: "AuthError" mid-session (token expired, refresh denied), your client's session lapsed — re-trigger its sign-in (your client re-prompts on the next call), then retry the original call ONCE. Don't hunt for an auth tool your harness doesn't expose. If the rejection names tenant_not_allowed, do not re-trigger sign-in, see Rules.

If sign-in itself errors (server unreachable, DCR fails): surface the message verbatim and file against OutSystems/outsystems-mcp. Don't speculate about server internals. One exception: an error saying the OAuth callback port is already in use is a local port conflict, not a server fault, so don't file it. See Troubleshooting at https://github.com/OutSystems/outsystems-mcp#troubleshooting

Tools at a glance

Discover the live catalog from the MCP server's tools/list; treat each tool's description + inputSchema as the source of truth. Don't rely on a hardcoded list — the set can change server-side. The tools group into these domains:

  • Apps — list and inspect applications, their references, and revision history; create new apps by asking Mentor in a mentor turn (see Mentor session round-trip): give the kind, the name and the portfolio, and ask for the kind's standard application template, as the ODC Studio new-app wizard uses, unless the user wants a blank shell. Which kinds Mentor can create depends on the tenant; when it can't create the one asked for, the run says which it can.
  • Context Service — seven typed, read-only lookups over a tenant's elements (entities, actions, screens, structures, roles, themes, connections).
  • Mentor — server-side OML editing as an async, multi-turn session. How the steps are split across tools varies by tenant, so read tools/list before the first mentor call.
  • Publish — compile edited OML into a new revision that lands in the tenant's Development environment, by asking Mentor to publish inside a mentor turn. There is no separate publish tool; see Caveats.
  • Deployments — promote builds across environments, roll back, run impact analyses.
  • External libraries — upload, publish, inspect, and fetch source for .NET libraries.
  • Environments — enumerate the tenant's environments.

Caveats

Cross-tool behaviors not expressible in a single per-tool description:

  • Promoting a build needs an explicit source environment. When you start a deployment with neither a build_key nor a revision pinned (and the operation isn't an undeploy), pass from_env; the target env_key is not used as the source on HTTP.
  • Revision notes attach on publish, not on promotion. A deployment operation carries no comment field of its own; the optional note on a publish is the only place one is written (ODC Studio's "1-Click Publish with message"). Promoting moves an existing revision, so the note set when that revision was published is the one that travels to the target environment — when a user asks for a commit -m style comment on a promotion, point them at the publish that created the revision rather than reporting it as unsupported.
  • Publish edited OML inside a mentor turn; there is no publish tool. The only way to publish edited OML is a mentor turn whose prompt asks Mentor to publish. The turn publishes the session's apps and takes no app-key argument; when the session holds more than one app, name the ones to publish in the prompt and pass their keys in the prompt's asset references. It lands in the development environment. It also reconciles a newer revision someone else published in the meantime. Put the publish note in that prompt (500 characters at most) and read the outcome from the finished run, as described under Mentor session round-trip. Don't search the catalog for a direct publish call or tell the user publishing is unsupported because none is listed.
  • Uploading an external library has a 50 MB decoded cap (~67 MB encoded zip_b64) and per-replica concurrency gating. Pre-flight rejects oversize payloads. The concurrency gate is not a queue: an upload arriving while all per-replica permits are held waits 5 seconds and is then rejected with Server is busy, retry shortly. Current servers return it as a CapacityError with data.code = "server_busy"; older servers may still return it as a ValidationError with the same message, where the category alone reads as fatal. Both forms are transient: retry with backoff instead of reporting a permanent failure.
  • The context lookups default owned_only to true when app is set, false tenant-wide. Pass owned_only: false with app to keep rows inherited from referenced libraries (OutSystemsUI, Charts, etc.).
  • Not every deployment serves the app model graph. On a deployment that does not serve it yet, context_graph is absent from tools/list, and a call to it by name returns data.code = "tool_unavailable_on_transport". Do not retry the call on that deployment. To describe the app, use the app's references for its dependencies and the per-type context lookups for its elements. If the app's references answer refs_fallback_unavailable, this deployment cannot list that app's dependencies yet. With fallbackReason not_indexed, do not retry the call. Empty per-type lookups then do not prove that the app has no elements, so tell the user that.
  • A context_graph match proves only that the app imports or references the element, not that it uses it. To find the apps that import or reference an entity, call context_search first, not one context_graph call per app. If no comma part of root matches, the call succeeds with an empty graph and a root_note. Do not report that answer as an error. The tool description gives the other rules of root.
  • A deployment-impact analysis needs an already-deployed asset. A deployment analysis (delete: false) analyses the revision held by the environment with the asset's most recent deployment, so an asset that has never been deployed fails until it is published; a publish to the development environment is enough. A Workflow, or an ExtensionLibrary, LowCodeLibrary, WidgetLibrary, MobileLibrary or ExternalLibrary asset, cannot be analysed this way even after a publish; before deploying one, still ask the user to confirm and tell them no impact analysis is available. A deletion analysis (delete: true) needs no deployed revision, and running one in place of a deployment analysis answers a different question.

Rules

  • Read tools/list before your first non-auth call. This skill names domains; the server names tools. (Auth comes first; see Authenticating.)
  • Platform first. This server is for OutSystems Developer Cloud (ODC). OutSystems 11 (O11) modules open in Service Studio use the Service Studio MCP server (named servicestudio in the O11 setup) instead. If a request is for work on O11, in an O11 module or the module open in Service Studio, don't use this server's tools, and if no Service Studio MCP tools are visible, tell the user to start the MCP server in Service Studio and stop. A request for work on ODC that only mentions O11, for example recreating an O11 entity in an ODC app, is for ODC. When both servers are connected, run the platform check at the top of this skill before your first call. Once the user names the platform, or earlier turns fixed it, treat later requests in this conversation as for that platform unless the user says otherwise.
  • Agent-facing tools. Don't expose raw tool output; extract the relevant fields and present them naturally.
  • Go straight to the task. No setup checks, no auth pre-flight beyond the lazy sign-in described above; identity comes from the harness-negotiated bearer. The platform check at the top of this skill is not a setup check: when both OutSystems MCP servers are connected, it comes before anything else.
  • Confirm before tenant-state mutations. Before invoking any tool that can change tenant state, restate the planned change to the user and wait for explicit confirmation. Skip the prompt only when the user h

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars0
CategoryDevelopment
Updated3d ago
Forks0

Trust signals

80/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 medium1 low
outsystems-mcp — Universal Skill: Install & Safety Check | SkillAgent