agents-connect
An agent skill by aws, from aws/agent-toolkit-for-aws. Tags: api, automation, integration, mcp, security.
What it does
Use when connecting your agent to external APIs, tools, or services via Gateway, or restricting tool access with Cedar policies. Handles gateway setup, target types, outbound auth (OAuth, API key, IAM), credentials, and Cedar policy authoring. Triggers on: "connect to API", "add gateway", "connect to MCP server", "Lambda tools", "OpenAPI", "gateway target", "Cedar policy", "restrict tools", "policy engine", "gateway auth error", "store API key", "outbound credential", "env var API key", "API key None after deploy", "credential not available after deploy", "should this be a gateway target", "give my agent tools", "add tools to agent". Not for inbound auth (who can call your agent) — use agents-harden. Not for debugging agent behavior — use agents-debug. Not for VPC networking errors (agent can't reach APIs due to VPC) — use agents-build. Not for creating or hosting a new MCP server project — use agents-get-started.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add aws/agent-toolkit-for-aws --skill agents-connect
Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):
git clone --depth 1 https://github.com/aws/agent-toolkit-for-aws
cp -r agent-toolkit-for-aws/plugins/aws-agents/skills/agents-connect ~/.claude/skills/agents-connect
Safety box score
Grade C, moderate risk: 40 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.
This skill provisions AgentCore gateways, gateway targets, credentials, and Cedar policies so an agent can call external services. The main risks are cloud-side deployment/change control, handling API keys or OAuth secrets, and misconfigurations that either expose tools publicly or block them entirely.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 2 (moderate) | It requires running operational CLI/Bash commands such as `agentcore --version`, `agentcore add gateway-target`, and `agentcore deploy -y`, so installing it enables command execution in the project workflow. |
| Network access | 2 (moderate) | Its purpose is to connect the agent to external MCP/OpenAPI/Lambda/API endpoints and send requests to remote services, e.g. `--endpoint https://mcp.example.com/mcp` and `https://api.example.com/endpoint`. |
| Filesystem reach | 1 (low) | It reads `agentcore/agentcore.json` and may use `agentcore/.env.local`, but the documented file access stays within the workspace and project config. |
| Secrets & credentials | 2 (moderate) | It handles credential material directly, including `--api-key`, `--client-secret`, and storing values in `agentcore/.env.local` for local dev. |
| Privilege & persistence | 0 (none) | No sudo, global install, service registration, shell startup editing, or other OS-level privilege changes are described. |
| Instruction hijack surface | 1 (low) | The docs repeatedly use `agentcore deploy -y`, which auto-confirms prompts, but they do not otherwise instruct the agent to ignore higher-level rules or hide information. |
| Irreversible actions | 2 (moderate) | It can deploy gateways and policy engines and switch them to `ENFORCE`, including deny-all Cedar policies, which materially changes who can call tools. |
| Transparency | 0 (none) | All shipped files are plain markdown; there are no binaries, minified blobs, or decoded-at-runtime code. |
Before installing
- Read `SKILL.md` sections for Path A/B/C/D and confirm whether you truly need direct credentials instead of a Gateway target.
- Inspect `agentcore/agentcore.json` for existing gateways and policy engines before running deploy.
- Review `references/policy.md` if you plan to use ENFORCE mode or generate Cedar policies.
- Run `agentcore add gateway-target --help` and `agentcore add credential --help` in a scratch project to confirm current auth flags and secret handling.
Rated 2026-10-20 by an automated review of 2 files from the skill's repository.
Source
- Repository
- aws/agent-toolkit-for-aws (all skills from this repository)
- Path
- plugins/aws-agents/skills/agents-connect/SKILL.md
- Branch
- main
- Collection
- aws-agents
- Updated
- 2026-09-19
Related skills
- setup — Set up the AWS DevOps Agent and AWS Security Agent connections. Use when the user says "set up", "configure", "connect", or when MCP tools are missing.
- setup-security-agent — Configure AWS Security Agent for the current workspace — provision or reuse an agent space, IAM service role, and S3 bucket.
- atlassian-mcp — Integrates with Atlassian products to manage project tracking and documentation via MCP protocol.
- fullstack-guardian — Builds security-focused full-stack web applications by implementing integrated frontend and backend components with layered security at every level.
- hyperframes-registry — Search, install, and wire registry blocks and components into HyperFrames compositions.
- mcp-developer — Use when building, debugging, or extending MCP servers or clients that connect AI systems with external tools and data sources.