Add OpenVLA AgentCore Orchestrator: autonomous Slurm training orchestration

[mijanur132]

Summary

Adds a new use case under 02-use-cases/ml-training-agent/ demonstrating how Amazon Bedrock AgentCore can autonomously manage GPU training jobs on a SageMaker HyperPod Slurm cluster.

What's included

File	Purpose
slurm_mcp_server.py	MCP server exposing 6 Slurm tools (submit, status, logs, cancel, info, metrics) over SSH
vla_training_agent.py	Autonomous agent with anomaly detection and recovery
openvla.Dockerfile	Training container with pinned deps (torch, transformers, peft) on EFA/NCCL base
slurm/finetune_openvla.sbatch	Batch script using srun --container-image + torchrun
.env.example	Configuration template
requirements.txt	MCP server dependency

Architecture

Same pattern as the SRE-agent sibling:

• MCP tools → Slurm operations (instead of K8s/logs/metrics APIs)
• Agent → Monitors training metrics (instead of infrastructure health)
• Recovery → Cancels diverging jobs, adjusts LR, resubmits (instead of pod restarts)

Concrete workload

OpenVLA-7B LoRA fine-tuning on LIBERO robotics benchmark. Tested on HyperPod P5en (8x H200): ~10 min for 500 steps, 15 GB checkpoint output.

Relation to awsome-distributed-ai PR

The standalone training recipe (Dockerfile + sbatch, no agent) is submitted separately as awslabs/awsome-distributed-ai#1112. This PR bundles the full agent + training workload as a self-contained AgentCore demo.

[github-actions]

Latest scan for commit: 977f929 | Updated: 2026-06-02 15:13:27 UTC

Security Scan Results

Scan Metadata

• Project: ASH
• Scan executed: 2026-06-02T15:13:06+00:00
• ASH version: 3.0.0

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	H	M	L	I	Time	Action	Result	Thresh
bandit	0	0	0	0	2	0	797ms	0	PASSED	MED (g)
cdk-nag	0	0	0	0	0	0	30.1s	0	PASSED	MED (g)
cfn-nag	0	0	0	0	0	0	34ms	0	PASSED	MED (g)
checkov	0	0	0	0	0	0	5.4s	0	PASSED	MED (g)
detect-secrets	0	0	0	0	0	0	760ms	0	PASSED	MED (g)
grype	0	0	0	0	0	0	46.0s	0	PASSED	MED (g)
npm-audit	0	0	0	0	0	0	205ms	0	PASSED	MED (g)
opengrep	0	0	0	0	0	0	<1ms	0	SKIPPED	MED (g)
semgrep	0	0	0	0	0	0	<1ms	0	MISSING	MED (g)
syft	0	0	0	0	0	0	2.5s	0	PASSED	MED (g)