Add AgentCore AWS Specialist Agent sample by ren8k · Pull Request #105 · aws-samples/sample-FAST-applications

ren8k · 2026-07-17T02:34:31Z

What this PR adds

A new sample under samples/aws-specialist-agent/: an AWS specialist chat agent built on FAST v0.4.1, showcased at the AgentCore booth at AWS Summit Japan 2026. The agent reasons about AWS, calls AWS APIs and managed tools through AgentCore Gateway, searches the web, executes code, and remembers facts across sessions.

Key differences from FAST

Selectable models: Claude and OpenAI GPT on Bedrock via a CDK model registry and per-chat UI selector
Long-term memory: AgentCore Memory SemanticMemoryStrategy plus an LTM-listing MCP server hosted on its own Runtime
Web search: Amazon-managed AgentCore Gateway web search connector target
Chat-history sidebar: session titles and history restored across logins (API Gateway + Lambda + DynamoDB)
Fully closed NAT-free VPC: the Runtime and auxiliary Lambdas egress through interface/gateway VPC endpoints only
Skills on Runtime: AWS skills mounted read-only from S3 Files at /mnt/skills, vendored and hash-pinned
Multiple MCP server management: remote MCP, Runtime-hosted MCP servers, a managed connector, and a Lambda target, gated per user department by Cedar ABAC
Speculative pre-warming: frontend pre-warm calls plus Runtime lifecycle configuration to cut cold-start TTFB

Checklist against CONTRIBUTING.md

Built on FAST v0.4.1; the (uksb-v6dos0t5g8) tracking suffix in infra-cdk/lib/fast-main-stack.ts is preserved
Descriptive kebab-case directory name and stack_name_base (fast-specialist-agent) in config.yaml (the name avoids the word "aws" because Cognito domain prefixes, derived from stack_name_base, must not contain the reserved words aws/amazon/cognito)
Top-level sample README with Overview / Key Differences / Architecture / Prerequisites / Deployment / Usage, an architecture diagram (drawio + png) modified from the FAST original, and a UI screenshot + demo gif in docs/img/
Only adds files under samples/aws-specialist-agent/ and appends one entry to the root README (table row + section, no other lines touched)
CDK is the only IaC option (no infra-terraform directory)
No customer data, datasets/models, hardcoded account IDs, real emails, internal URLs, or deployment-specific values; deployment config uses committed placeholders
Security review completed through the internal process for FAST samples; ASH scan is green (suppressions documented in .ash/ash.yaml)
Tests and lint pass: pytest 60/60, vitest 102/102, CDK jest 18/18, ruff/eslint/prettier clean
Deployment verified end to end from this config: the stack reaches CREATE_COMPLETE and all Cedar policies are created ACTIVE

An AWS specialist chat agent built on FAST v0.4.1, showcased at the AgentCore booth at AWS Summit Japan 2026. Key additions over the FAST baseline: - Selectable models: Claude and OpenAI GPT on Bedrock via a CDK model registry and per-chat UI selector - Long-term memory: AgentCore Memory SemanticMemoryStrategy plus an LTM-listing MCP server hosted on its own Runtime - Web search: Amazon-managed AgentCore Gateway web search connector - Chat-history sidebar: API Gateway + Lambda + DynamoDB session history - Fully closed NAT-free VPC: all egress via interface/gateway endpoints - Skills on Runtime: read-only S3 Files mount at /mnt/skills with vendored, hash-pinned skill content - Multi-MCP-server management with Cedar ABAC per user department - Speculative pre-warming to reduce cold-start latency ASH scan green with suppressions documented in .ash/ash.yaml. Tests: pytest 57/57, vitest 102/102, CDK jest 18/18.

github-actions · 2026-07-17T02:39:15Z

Latest scan for commit: 45630ee | Updated: 2026-07-17 16:10:20 UTC

Security Scan Results

Scan Metadata

Project: ASH
Scan executed: 2026-07-17T16:10:07+00:00
ASH version: 3.2.2

Summary

Scanner Results

The table below shows findings by scanner, with status based on severity thresholds and dependencies:

Column Explanations:

Severity Levels (S/C/H/M/L/I):

Suppressed (S): Security findings that have been explicitly suppressed/ignored and don't affect the scanner's pass/fail status
Critical (C): The most severe security vulnerabilities requiring immediate remediation (e.g., SQL injection, remote code execution)
High (H): Serious security vulnerabilities that should be addressed promptly (e.g., authentication bypasses, privilege escalation)
Medium (M): Moderate security risks that should be addressed in normal development cycles (e.g., weak encryption, input validation issues)
Low (L): Minor security concerns with limited impact (e.g., information disclosure, weak recommendations)
Info (I): Informational findings for awareness with minimal security risk (e.g., code quality suggestions, best practice recommendations)

Other Columns:

Time: Duration taken by each scanner to complete its analysis
Action: Total number of actionable findings at or above the configured severity threshold that require attention

Scanner Results:

PASSED: Scanner found no security issues at or above the configured severity threshold - code is clean for this scanner
FAILED: Scanner found security vulnerabilities at or above the threshold that require attention and remediation
MISSING: Scanner could not run because required dependencies/tools are not installed or available
SKIPPED: Scanner was intentionally disabled or excluded from this scan
ERROR: Scanner encountered an execution error and could not complete successfully

Severity Thresholds (Thresh Column):

CRITICAL: Only Critical severity findings cause scanner to fail
HIGH: High and Critical severity findings cause scanner to fail
MEDIUM (MED): Medium, High, and Critical severity findings cause scanner to fail
LOW: Low, Medium, High, and Critical severity findings cause scanner to fail
ALL: Any finding of any severity level causes scanner to fail

Threshold Source: Values in parentheses indicate where the threshold is configured:

(g) = global: Set in the global_settings section of ASH configuration
(c) = config: Set in the individual scanner configuration section
(s) = scanner: Default threshold built into the scanner itself

Statistics calculation:

All statistics are calculated from the final aggregated SARIF report
Suppressed findings are counted separately and do not contribute to actionable findings
Scanner status is determined by comparing actionable findings to the threshold

Scanner	S	C	M	L	Time	Action	Result	Thresh
bandit	0	4	0	69	1.2s	4	FAILED	MED (g)
cdk-nag	0	0	0	0	13.7s	0	PASSED	MED (g)
cfn-nag	0	0	0	0	1.6s	0	PASSED	MED (g)
checkov	0	0	0	0	7.6s	0	PASSED	MED (g)
detect-secre…	0	6	0	0	5.5s	6	FAILED	MED (g)
grype	0	9	1	0	1m 0s	10	FAILED	MED (g)
npm-audit	0	0	0	0	1.4s	0	PASSED	MED (g)
opengrep	12	6	0	0	22.5s	6	FAILED	MED (g)
semgrep	0	0	0	0	<1ms	0	MISSING	MED (g)
syft	0	0	0	0	2.9s	0	PASSED	MED (g)

Detailed Findings

Show 26 actionable findings

Finding 1: B104

Severity: HIGH
Scanner: bandit
Rule ID: B104
Location: samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/server.py:42-44

Description:
Possible binding to all interfaces.

Code Snippet:

mcp = FastMCP("ltm", host="0.0.0.0", stateless_http=True)

Finding 2: B104

Severity: HIGH
Scanner: bandit
Rule ID: B104
Location: samples/aws-specialist-agent/gateway/tools/strands_mcp_server/server.py:35-37

Description:
Possible binding to all interfaces.

Code Snippet:

# AgentCore Runtime MCP contract.
mcp.settings.host = "0.0.0.0"
mcp.settings.port = 8000

Finding 3: B104

Severity: HIGH
Scanner: bandit
Rule ID: B104
Location: samples/aws-specialist-agent/tests/unit/test_ltm_mcp_server.py:140-142

Description:
Possible binding to all interfaces.

Code Snippet:

_load_server_module()
    assert captured["fastmcp_kwargs"] == {"host": "0.0.0.0", "stateless_http": True}

Finding 4: B104

Severity: HIGH
Scanner: bandit
Rule ID: B104
Location: samples/aws-specialist-agent/tests/unit/test_strands_mcp_server.py:101-103

Description:
Possible binding to all interfaces.

Code Snippet:

mod = _load_server_module()
    assert mod.mcp.settings.host == "0.0.0.0"
    assert mod.mcp.settings.port == 8000

Finding 5: SECRET-HEX-HIGH-ENTROPY-STRING

Severity: HIGH
Scanner: detect-secrets
Rule ID: SECRET-HEX-HIGH-ENTROPY-STRING
Location: samples/aws-specialist-agent/scripts/vendor-skills.py:30

Description:
Secret of type 'Hex High Entropy String' detected in file 'samples/aws-specialist-agent/scripts/vendor-skills.py' at line 30

Code Snippet:

Secret of type Hex High Entropy String detected

Finding 6: SECRET-HEX-HIGH-ENTROPY-STRING

Severity: HIGH
Scanner: detect-secrets
Rule ID: SECRET-HEX-HIGH-ENTROPY-STRING
Location: samples/aws-specialist-agent/skills-lock.json:8

Description:
Secret of type 'Hex High Entropy String' detected in file 'samples/aws-specialist-agent/skills-lock.json' at line 8

Code Snippet:

Secret of type Hex High Entropy String detected

Finding 7: SECRET-HEX-HIGH-ENTROPY-STRING

Severity: HIGH
Scanner: detect-secrets
Rule ID: SECRET-HEX-HIGH-ENTROPY-STRING
Location: samples/aws-specialist-agent/skills-lock.json:14

Description:
Secret of type 'Hex High Entropy String' detected in file 'samples/aws-specialist-agent/skills-lock.json' at line 14

Code Snippet:

Secret of type Hex High Entropy String detected

Finding 8: SECRET-SECRET-KEYWORD

Severity: HIGH
Scanner: detect-secrets
Rule ID: SECRET-SECRET-KEYWORD
Location: samples/aws-specialist-agent/test-scripts/test-runtime-vpc-e2e.py:61

Description:
Secret of type 'Secret Keyword' detected in file 'samples/aws-specialist-agent/test-scripts/test-runtime-vpc-e2e.py' at line 61

Code Snippet:

Secret of type Secret Keyword detected

Finding 9: SECRET-SECRET-KEYWORD

Severity: HIGH
Scanner: detect-secrets
Rule ID: SECRET-SECRET-KEYWORD
Location: samples/aws-specialist-agent/test-scripts/test-skills-e2e.py:58

Description:
Secret of type 'Secret Keyword' detected in file 'samples/aws-specialist-agent/test-scripts/test-skills-e2e.py' at line 58

Code Snippet:

Secret of type Secret Keyword detected

Finding 10: SECRET-SECRET-KEYWORD

Severity: HIGH
Scanner: detect-secrets
Rule ID: SECRET-SECRET-KEYWORD
Location: samples/aws-specialist-agent/tests/unit/test_models.py:229

Description:
Secret of type 'Secret Keyword' detected in file 'samples/aws-specialist-agent/tests/unit/test_models.py' at line 229

Code Snippet:

Secret of type Secret Keyword detected

Finding 11: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Severity: HIGH
Scanner: opengrep
Rule ID: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure
Location: samples/aws-specialist-agent/agent/utils/auth.py:215-217

Description:
Detected a python logger call with a potential hardcoded secret "Getting access token for stack: %s, region: %s" being logged. This may lead to secret credentials being exposed. Make sure that the logger is not logging sensitive information.

Code Snippet:

logger.info(
        "Getting access token for stack: %s, region: %s", stack_name, region
    )  # nosemgrep: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Finding 12: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Severity: HIGH
Scanner: opengrep
Rule ID: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure
Location: samples/aws-specialist-agent/agent/utils/auth.py:250-252

Description:
Detected a python logger call with a potential hardcoded secret "Requesting token from: %s" being logged. This may lead to secret credentials being exposed. Make sure that the logger is not logging sensitive information.

Code Snippet:

logger.info(
        "Requesting token from: %s", token_url
    )  # nosemgrep: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Finding 13: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Severity: HIGH
Scanner: opengrep
Rule ID: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure
Location: samples/aws-specialist-agent/agent/utils/auth.py:259-261

Description:
Detected a python logger call with a potential hardcoded secret "Token request failed: %s" being logged. This may lead to secret credentials being exposed. Make sure that the logger is not logging sensitive information.

Code Snippet:

logger.error(
            "Token request failed: %s", response.status_code
        )  # nosemgrep: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Finding 14: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal

Severity: HIGH
Scanner: opengrep
Rule ID: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal
Location: samples/aws-specialist-agent/infra-cdk/lib/backend-stack.ts:232

Description:
Detected possible user input going into a path.join or path.resolve function. This could possibly lead to a path traversal vulnerability, where the attacker can access arbitrary files stored in the file system. Instead, be sure to sanitize or validate user input first.

Code Snippet:

? path.join(prefix, entry.name)

Finding 15: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal

Severity: HIGH
Scanner: opengrep
Rule ID: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal
Location: samples/aws-specialist-agent/infra-cdk/lib/backend-stack.ts:232

Description:
Detected possible user input going into a path.join or path.resolve function. This could possibly lead to a path traversal vulnerability, where the attacker can access arbitrary files stored in the file system. Instead, be sure to sanitize or validate user input first.

Code Snippet:

? path.join(prefix, entry.name)

Finding 16: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal

Severity: HIGH
Scanner: opengrep
Rule ID: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal
Location: samples/aws-specialist-agent/infra-cdk/lib/backend-stack.ts:1978

Description:
Detected possible user input going into a path.join or path.resolve function. This could possibly lead to a path traversal vulnerability, where the attacker can access arbitrary files stored in the file system. Instead, be sure to sanitize or validate user input first.

Code Snippet:

.readFileSync(path.join(policiesDir, f), "utf-8")

Finding 17: GHSA-jpw9-pfvf-9f58-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-jpw9-pfvf-9f58-mcp
Location: samples/aws-specialist-agent/agent/strands-single-agent/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/agent/strands-single-agent/requirements.txt

Finding 18: GHSA-jpw9-pfvf-9f58-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-jpw9-pfvf-9f58-mcp
Location: samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/requirements.txt

Finding 19: GHSA-jpw9-pfvf-9f58-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-jpw9-pfvf-9f58-mcp
Location: samples/aws-specialist-agent/gateway/tools/strands_mcp_server/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/gateway/tools/strands_mcp_server/requirements.txt

Finding 20: GHSA-hvrp-rf83-w775-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-hvrp-rf83-w775-mcp
Location: samples/aws-specialist-agent/agent/strands-single-agent/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/agent/strands-single-agent/requirements.txt

Finding 21: GHSA-hvrp-rf83-w775-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-hvrp-rf83-w775-mcp
Location: samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/requirements.txt

Finding 22: GHSA-hvrp-rf83-w775-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-hvrp-rf83-w775-mcp
Location: samples/aws-specialist-agent/gateway/tools/strands_mcp_server/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/gateway/tools/strands_mcp_server/requirements.txt

Finding 23: GHSA-vj7q-gjh5-988w-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-vj7q-gjh5-988w-mcp
Location: samples/aws-specialist-agent/agent/strands-single-agent/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/agent/strands-single-agent/requirements.txt

Finding 24: GHSA-vj7q-gjh5-988w-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-vj7q-gjh5-988w-mcp
Location: samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/gateway/tools/ltm_mcp_server/requirements.txt

Finding 25: GHSA-vj7q-gjh5-988w-mcp

Severity: HIGH
Scanner: grype
Rule ID: GHSA-vj7q-gjh5-988w-mcp
Location: samples/aws-specialist-agent/gateway/tools/strands_mcp_server/requirements.txt:1

Description:
A high vulnerability in python package: mcp, version 1.27.1 was found at: /samples/aws-specialist-agent/gateway/tools/strands_mcp_server/requirements.txt

Finding 26: GHSA-gh4j-gqv2-49f6-fast-xml-parser

Severity: MEDIUM
Scanner: grype
Rule ID: GHSA-gh4j-gqv2-49f6-fast-xml-parser
Location: samples/aws-specialist-agent/infra-cdk/package-lock.json:1

Description:
A medium vulnerability in npm package: fast-xml-parser, version 5.5.8 was found at: /samples/aws-specialist-agent/infra-cdk/package-lock.json

Report generated by Automated Security Helper (ASH) at 2026-07-17T16:10:03+00:00

…ncation Deployment verification of the sample surfaced two failures: - Cognito domain prefixes must not contain the reserved words aws/amazon/ cognito, and cognito-stack.ts derives the prefix from stack_name_base, so 'fast-aws-specialist' failed UserPoolDomain creation with InvalidRequest. Rename to 'fast-specialist-agent'. - The cedar-policy Lambda truncated over-long policy names with policy_name[:48], which drops the index suffix and gives every policy the same name once the engine name exceeds ~30 chars; CreatePolicy then fails with ConflictException. Shorten the timestamp instead and always keep the index. Verified end to end: the stack deploys CREATE_COMPLETE and all six Cedar policies are created ACTIVE with unique names.

…sion test

Show the agent invoking AWS Skills, AWS MCP tools, and managed Web Search via AgentCore Gateway while rendering a mermaid architecture diagram, with red callouts for the chat-history sidebar and the model selector. Addresses PR aws-samples#105 review feedback to better differentiate the sample from out-of-the-box FAST.

github-actions Bot added documentation Improvements or additions to documentation infrastructure frontend backend docker tests labels Jul 17, 2026

ren8k added 4 commits July 17, 2026 13:21

docs: link the Zenn deep-dive article, add Cedar policy naming regres…

4956758

…sion test

docs: use Google-translated Zenn URL in the English README

bb60b98

kaleko approved these changes Jul 17, 2026

View reviewed changes

kaleko merged commit f14a56d into aws-samples:main Jul 17, 2026
5 checks passed

Uh oh!

Conversation

ren8k commented Jul 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What this PR adds

Key differences from FAST

Checklist against CONTRIBUTING.md

Uh oh!

github-actions Bot commented Jul 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Security Scan Results

Scan Metadata

Summary

Scanner Results

Detailed Findings

Finding 1: B104

Finding 2: B104

Finding 3: B104

Finding 4: B104

Finding 5: SECRET-HEX-HIGH-ENTROPY-STRING

Finding 6: SECRET-HEX-HIGH-ENTROPY-STRING

Finding 7: SECRET-HEX-HIGH-ENTROPY-STRING

Finding 8: SECRET-SECRET-KEYWORD

Finding 9: SECRET-SECRET-KEYWORD

Finding 10: SECRET-SECRET-KEYWORD

Finding 11: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Finding 12: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Finding 13: python.lang.security.audit.logging.logger-credential-leak.python-logger-credential-disclosure

Finding 14: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal

Finding 15: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal

Finding 16: javascript.lang.security.audit.path-traversal.path-join-resolve-traversal.path-join-resolve-traversal

Finding 17: GHSA-jpw9-pfvf-9f58-mcp

Finding 18: GHSA-jpw9-pfvf-9f58-mcp

Finding 19: GHSA-jpw9-pfvf-9f58-mcp

Finding 20: GHSA-hvrp-rf83-w775-mcp

Finding 21: GHSA-hvrp-rf83-w775-mcp

Finding 22: GHSA-hvrp-rf83-w775-mcp

Finding 23: GHSA-vj7q-gjh5-988w-mcp

Finding 24: GHSA-vj7q-gjh5-988w-mcp

Finding 25: GHSA-vj7q-gjh5-988w-mcp

Finding 26: GHSA-gh4j-gqv2-49f6-fast-xml-parser

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

ren8k commented Jul 17, 2026 •

edited

Loading

github-actions Bot commented Jul 17, 2026 •

edited

Loading