Skip to content
Agentmart
Submit a tool
This entry duplicates adding-an-mcp-tool, which is the one we list in search.

adding-an-mcp-tool

Use when adding a new MCP tool or capability to the AI Orchestration server (a new tool under modules/mcp/server).

Skill by ufukyuksel-dev

IndexedMIT
Write a review

Chalk Labs

Need help deploying adding-an-mcp-tool?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/ufukyuksel-dev-adding-an-mcp-tool-a41611

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when adding a new MCP tool or capability to the AI Orchestration server (a new tool under modules/mcp/server).

Use cases: backend, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions: not checked yet.

What's inside

Skill name
adding-an-mcp-tool
Description
Use when adding a new MCP tool or capability to the AI Orchestration server (a new tool under modules/mcp/server).
License
MIT
Instructions
19 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Adding an MCP tool

> This is about **building** a tool on the AI Orchestration server. It is not
> permission for Copilot to call this service's MCP tools natively: Copilot's
> only interface remains the `ai_orch` terminal command. Do not rename the
> server's MCP architecture just to remove the word "MCP".

1. **Ambiguity stop:** if the requested tool name, scope, project boundary, or side effect is unclear after two focused passes, ask the user before encoding a permanent MCP contract. Give 2-3 likely interpretations and the evidence for each.
2. Create `XxxMcpTool` in `modules/mcp/server`: `@Component` + `@ConditionalOnProperty(prefix="ai-orchestration.mcp", name="enabled", havingValue="true", matchIfMissing=true)`. Methods are annotated `@McpTool(name="x.y", description=...)` with `@McpToolParam` per argument.
3. In each method: resolve context via `McpClientContextHolder.require()`, **scope-check** (`requireScope`/`context.hasScope`), and **audit** via `McpAuditLogger` for every outcome — success, `denied_scope`, error. Reject cross-project access.
4. Register the scope in **all** of: `McpApiKeyService` tiers (prod keys); `LocalTrustProperties.DEFAULT_LOCAL_SCOPES` (fallback); the `ai-orchestration.local-trust.local-scopes` list in **both** `application.yml` and `application-local-first.yml` (these YAML lists override the constant, so the scope won't apply in local-first without them); and a Flyway `Vn__<name>_scope.sql` granting it to existing keys.
5. Never bypass human-approval gates (hard-delete params, `memory.confirm`); redact before any external egress.
6. Tests: missing-scope rejected + audited; happy path; ideally an integration test asserting the tool registers.

7. If the new tool should also be reachable from Copilot, add it deliberately to the `ai_orch` allowlist, argument table, scope class and command builder in `skills/copilot/ai-orchestration-memory/scripts/ai_orch.py`, with tests and a row in `docs/agent/copilot-capability-matrix.md`. A tool existing on the server never means Copilot may call it: there is no generic dispatcher and the local allowlist is not widened by the server catalog.

Mirror an existing tool: `MemoryMcpTool`, `CodebaseMcpTool`, or `ScannerMcpTool`.

From ufukyuksel-dev/ai-orchestration, MIT.

More from ufukyuksel-dev/ai-orchestration

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 27, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

Same skill elsewhere

Copies of this SKILL.md published by others; search shows only one of them.

More from ufukyuksel-dev

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get adding-an-mcp-tool running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying adding-an-mcp-tool?

CallGet help
Cite this page

adding-an-mcp-tool skill. Agent Mart. https://agentmart.dev/skills/ufukyuksel-dev-adding-an-mcp-tool-a41611 (last updated 2026-10-24)