CtrlK
BlogDocsLog inGet started
Tessl Logo

agents-connect

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.

72

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable and well-structured with strong progressive disclosure, offering executable code and a clear multi-path workflow. Its main weakness is repetition of the same auth caveats across the matrix, Path A, and troubleshooting sections, which slightly hurts token efficiency.

Suggestions

Consolidate the 2LO-vs-3LO OAuth and the 'MCP targets don't support API key' caveats into a single canonical statement in the auth matrix, then link from Path A and Troubleshooting rather than re-explaining each time.

Tighten the 'Before you reach for Path D' list (lines 437-442) since the same five direct-call situations are already tabulated at lines 102-108 — a back-reference would save tokens.

Add an explicit post-deploy verification checkpoint (e.g., list_tools() returns the expected tool) to close the workflow loop and lift workflow_clarity to 5.

DimensionReasoningScore

Conciseness

The body is mostly efficient and assumes Claude's competence (no basic-concept padding), but key facts are restated across sections — e.g. 'CLI only does 2LO OAuth' appears in the auth matrix (line 178), Path A (lines 268-272), and troubleshooting (lines 532-533), and 'MCP targets don't support API key' repeats at lines 182, 272, and 530 — which could be tightened.

3 / 5

Actionability

Provides copy-paste-ready, executable CLI commands and complete Python examples for each path (framework-agnostic MCP client, Strands, LangGraph, @requires_api_key/@requires_access_token decorators), covering the common cases including the local-dev-gap guard.

5 / 5

Workflow Clarity

A clearly sequenced Process (Step 0 version check → Step 1 read project → Step 2 identify target → Step 3 auth matrix → Paths A-D) with most checkpoints present (version-gate remediation, GATEWAY_URL guard before use, post-deploy 'fetch access'), but no explicit post-deploy verification loop confirming the tool is discoverable.

4 / 5

Progressive Disclosure

The body is an overview that points to one-level-deep, clearly-signaled references at the right moment — notably the confirmed-real references/policy.md (line 41) for Cedar details — with content appropriately split and easy to navigate.

5 / 5

Total

17

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is a strong, trigger-rich statement that clearly defines the skill's niche and when to invoke it, while explicitly distinguishing it from adjacent skills. It earns top marks across all four dimensions.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities — 'gateway setup, target types, outbound auth (OAuth, API key, IAM), credentials, and Cedar policy authoring' — giving comprehensive coverage rather than a single generic action.

5 / 5

Completeness

Explicitly answers both 'what' (the capability list) and 'when' via a clear 'Use when...' clause and a 'Triggers on:' list of concrete trigger phrases.

5 / 5

Trigger Term Quality

Includes an extensive set of natural user phrasings ('connect to API', 'add gateway', 'connect to MCP server', 'restrict tools', 'give my agent tools') plus error-message triggers ('API key None after deploy', 'credential not available after deploy'), covering synonyms and file-format terms.

5 / 5

Distinctiveness Conflict Risk

Has a clear niche (AgentCore Gateway + Cedar policy) and explicitly carves out sibling skills ('Not for inbound auth — use agents-harden', 'Not for debugging — use agents-debug', etc.), minimizing overlap risk.

5 / 5

Total

20

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (557 lines); consider splitting into references/ and linking

Warning

relative_links

Relative link issues: 4 suspicious

Warning

Total

14

/

16

Passed

Repository
aws/agent-toolkit-for-aws
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.