{"name":"context-engineering: Build Governed AI Workflows on Amazon Bedrock","description":"A Python reference implementation of context engineering using Anthropic Claude and Amazon Titan on Amazon Bedrock. It shows how to retrieve organizational decisions, guide AI-generated work, and check that outputs follow the supplied context.","github":"https://github.com/outcomeops/context-engineering","url":"https://osrepos.com/repo/outcomeops-context-engineering","source":"osrepos.com","sourceDescription":"This repository profile is provided by osrepos.com, an open source repository discovery platform.","repositoryProfile":"https://osrepos.com/repo/outcomeops-context-engineering","generatedFor":"open source discovery and AI-assisted research","markdown":"https://osrepos.com/repo/outcomeops-context-engineering.md","json":"https://osrepos.com/repo/outcomeops-context-engineering.json","topics":["python","ai","rag","developer-tools","context-engineering","architecture-decision-records"],"keywords":["python","ai","rag","developer-tools","context-engineering","architecture-decision-records"],"stars":null,"summary":"A Python reference implementation of context engineering using Anthropic Claude and Amazon Titan on Amazon Bedrock. It shows how to retrieve organizational decisions, guide AI-generated work, and check that outputs follow the supplied context.","content":"## Overview\n\nOutcomeOps’ context-engineering repository is a Python reference implementation built on Amazon Bedrock, using Anthropic Claude for generation and Amazon Titan for embeddings. It demonstrates how to give AI systems relevant organizational knowledge, such as architecture decision records (ADRs), so their outputs reflect local standards rather than generic patterns.\n\nThe project frames context engineering as more than retrieval: it connects a corpus and retrieval pipeline to context injection, reviewable outputs, and enforcement. The examples use Spring PetClinic and ADRs, making this a practical reference for teams evaluating how to make AI-generated software work more consistent with their decisions.\n\n## Key Features\n\n- Implements a five-part model: corpus, retrieval, injection, output, and enforcement.\n- Uses ADRs and a Spring PetClinic codebase as a shared example corpus.\n- Runs examples with Anthropic Claude and Amazon Titan through Amazon Bedrock.\n- Includes comparisons of context-engineered output with output produced without context, and discusses differences from RAG and other tools.\n- Provides runnable examples organized by component, each with its own README and requirements file.\n- Includes case studies and an example of an A/B comparison where ADRs change generated code.\n\n## Use Cases\n\n- An engineering team can prototype ways to supply architecture decisions to an AI coding workflow before applying the pattern to its own corpus.\n- A platform or AI engineering team can examine how retrieval, model input, output review, and enforcement fit together beyond a basic RAG pipeline.\n- Technical leaders evaluating context-engineering claims can use the examples to inspect the approach and compare it with alternatives.\n- Teams already using Amazon Bedrock can adapt the examples to test whether organizational standards improve generated work.\n\n## Our Take\n\nThis is a clearly framed, hands-on reference for context engineering, but its quiet maintenance and single-contributor history make it better suited to evaluation and adaptation than as a dependency to rely on without review.\n\n**Good fit if:**\n- You want a concrete Python example of context engineering built on Amazon Bedrock, Claude, and Titan.\n- Your team has ADRs or similar organizational material and wants to explore using it to shape AI-generated outputs.\n- You are comfortable reviewing and adapting a reference implementation under the MIT license.\n\n**Look elsewhere if:**\n- You need a currently active project with recent commits or releases; the repository has had no commits in the last 90 days and no release in over a year.\n- You need a maintained project with a broader contributor base; all recorded contributions come from one person.\n- You cannot use Amazon Bedrock or do not want to configure AWS credentials and complete Anthropic’s one-time account access step.\n\n## Project Health\n\n| Signal | Value |\n|---|---|\n| Status | **Quiet**: no commits in the last 90 days and no release in a year |\n| Last commit | 2026-05-17 (5 months ago) |\n| Commits, last 90 days | 0 |\n| Releases, last 12 months | 0 |\n| Contributors | 1 |\n| Issues closed, last 90 days | 0 |\n| Pull requests merged, last 90 days | 0 |\n| Project age | 6 months |\n\nChecked on 2026-10-17 with the GitHub API.\n\n## Project Facts\n\n- Language: Python\n- License: MIT\n- Stars: 128\n- Forks: 8\n- Archived: no\n\n## What You Need\n\nDetected in the repository:\n\n- Python (main language on GitHub)\n\n## Getting Started\n\nThe examples require Python 3.11 or newer and an AWS account with credentials configured. From the repository root:\n\n```bash\ncd 01-corpus\npip install -r requirements.txt\npython ingest_adrs.py ./sample-adrs\n```\n\nSee the [README](https://github.com/outcomeops/context-engineering) and the component folders for setup details and additional commands.\n\n## License in Plain Words\n\n**MIT** (permissive).\n\n- Commercial use: yes\n- Modify and redistribute: yes\n- You must keep: the copyright and license notice\n- Share your changes: no\n\nA summary, not legal advice: the LICENSE file is what applies.\n\n## Alternatives\n\n- [ace](https://osrepos.com/repo/ace-agent-ace): ACE builds reusable agent playbooks from task outcomes, while context-engineering retrieves organizational decisions and checks generated work against them.\n\n| Project | Language | License | Stars | Status |\n|---|---|---|---|---|\n| **context-engineering** | Python | MIT | 128 | Quiet |\n| [ace](https://osrepos.com/repo/ace-agent-ace) | Python | Apache-2.0 | 1.4k | Maintained |\n\n## Considerations\n\n- Examples depend on Amazon Bedrock, AWS credentials, and a supported region. Anthropic Claude also requires a one-time First Time Use form for an AWS account that has not used Anthropic models on Bedrock.\n- The project is marked quiet: it has no commits in the last 90 days and no release in over a year. It has one recorded contributor, so adopters should assess maintenance risk and verify behavior before relying on it.\n- The supplied project facts report no CI or tests. The repository is presented as a working reference implementation, not as a packaged service with those checks reported.\n- The MIT license permits broad reuse, subject to its license terms.","metrics":{"detailViews":1,"githubClicks":0},"dates":{"published":null,"modified":"2026-10-17T12:00:24.000Z"}}