7 views
This guide focuses on AI model provenance for production systems, with practical trade-offs for reliability, security, and cost.
Your coding agent image is part of the supply chain. If ECS/Lambda pulls :latest from a mutable tag with no attestation, you are running strangers’ code next to your repos. Aim for SLSA-ish provenance: signed images, pinned digests, verified before run.
⚡ TL;DR: Build in CI, sign with Sigstore/Notation, pin digests in task defs, verify admission. Track model IDs/versions the same way.
AI model provenance: production guidance
json
{
"image": "123.dkr.ecr.us-east-1.amazonaws.com/agent@sha256:4f2c…"
}
bash
# ✅ verify before deploy
cosign verify \
--certificate-identity-regexp 'CI' \
--certificate-oidc-issuer 'https://token.actions.githubusercontent.com' \
123.dkr.ecr.us-east-1.amazonaws.com/agent@sha256:4f2c…
Model provenance
Record provider, model ID, and inference profile ARN in the run audit. Refuse runs when the model ID is not in an allowlist.python
❌ Floating tags + mystery fine-tunes from a personal HF account in prod agents.
ALLOW_MODELS = {
"anthropic.claude-sonnet-4-20250514-v1:0",
"amazon.nova-pro-v1:0",
}
def assert_model(mid: str):
if mid not in ALLOW_MODELS:
raise RuntimeError(f"model_not_allowed:{mid}")
Closing checklist
- [ ] Digest-pinned images
- [ ] Signature verification in deploy
- [ ] Model allowlists
- [ ] SBOM for agent images
- [ ] Audit image digest per run
Series navigation
Day 66: Audit Trails Humans Can Replay · Day 68: Abuse and Cost Attacks Last updated September 11, 2026 Related: Day 76: p99 of Agents: Queueing, Not Just Model Latency; Day 78: Load Shedding When the Model Is Sick; Day 92: Threat Model an AI Feature.Watch: 100 Days of AI on YouTube
Short videos from CheatCoders, one AI engineering topic per day.
Most viewed
- Python Decorators Explained: From Simple Wrappers to Production Patterns
- Claude Code vs Cursor vs GitHub Copilot: Choose by Workflow, and Set Up One Repo That Works With All Three
- AI Agent Frameworks in 2025: LangGraph vs CrewAI vs AutoGen vs Raw API
- Python String Methods: Every str Method With Real Production Examples
- Css Animation Performance: Performance-Optimized Motion for Modern UIs
Newly added
- AWS Systems Manager Parameter Store: Hierarchical Runtime Config for Coding Agents
- Amazon Cognito Identity Pools: Federate Coding-Agent Tool Callers Without Long-Lived Keys
- Amazon Athena for Coding Agents: Query Tool Audit Logs and Eval Traces on S3 Without a Warehouse
- AWS Batch for Coding Agents: Parallel Overnight PR Reviews and Long-Running Jobs Without Lambda Timeouts
- Amazon Bedrock Prompt Management: Version and A/B Test Coding-Agent System Prompts
Deep-dive PDF
Get the expanded guide for this post — extra diagrams-style checklists, failure modes, and a production walkthrough. Free when you subscribe to CheatCoders.
Already subscribed? or open the subscribe page.
Discover more from CheatCoders
Subscribe to get the latest posts sent to your email.
