ContextForge RAG - Hermes Profile Setup¶
Source: codegraphtheory/context-forge-rag Stars: 1 ⭐ | License: MIT Created: June 24, 2026 Category: Agent Infrastructure / RAG / AI Architecture Type: Hermes Agent Profile (installable)
1. What It Is¶
ContextForge RAG is an installable Hermes Agent profile - not just a single skill, but a full agent configuration bundle for principal-level production RAG architecture. Built from codegraphtheory/hermes-profile-template (repo no longer public).
What this profile is good at: - Production RAG architecture: ingestion, chunking, embeddings, hybrid retrieval, reranking, citation packing, and model routing - Agent workflow design: intent classification, tool contracts, idempotency, escalation, bounded loops, failure handling - Evaluation & observability: golden datasets, retrieval metrics, LLM judge rubrics, traces, cost telemetry, regression gates - Delivery planning: ADRs, implementation-ready GitHub issues, roadmaps, rollout plans, rollback criteria
Bundled Skills¶
| Skill | Purpose |
|---|---|
principal-ai-architect |
ADRs, architecture tradeoffs, diagrams, decision records |
technical-roadmapping |
Delivery sequencing, milestones, dependency mapping, risk gates |
production-rag-architecture |
Retrieval system design, namespace strategy, chunking, embeddings, reranking, rollout |
rag-evaluation-observability |
Golden datasets, metric design, LLM judge rubrics, traces, regression gates |
agentic-workflow-issue-factory |
Implementation-ready issues and typed agent tool design |
2. Prerequisites¶
- Hermes Agent (profile install support)
- Any Hermes-supported model provider (interactive use)
- Optional:
.envvariables for external services (see.env.EXAMPLE)
3. Installation¶
Option A: Direct from GitHub¶
hermes profile install github.com/codegraphtheory/context-forge-rag --name context-forge-rag --yes
hermes -p context-forge-rag
Option B: Local Clone¶
git clone https://github.com/codegraphtheory/context-forge-rag.git
cd context-forge-rag
python3 scripts/validate_profile.py .
hermes profile install . --name context-forge-rag --yes
hermes -p context-forge-rag
Verify¶
python3 scripts/validate_profile.py .
python3 -m py_compile scripts/*.py
Generator Smoke Test¶
out=$(mktemp -d /tmp/context-forge-rag.XXXXXX)
python3 scripts/generate_profile.py --params templates/profile.params.yaml --output "$out"
python3 "$out/scripts/validate_profile.py" "$out"
4. First Prompts¶
Once the profile is active, these are productive starting points:
Design a production RAG architecture for customer support knowledge retrieval.
Include ingestion, metadata, evals, observability, and rollout gates.
Turn this ambiguous AI feature into 6 implementation-ready GitHub issues
with acceptance criteria and eval requirements.
Review my chunking and namespace strategy. Identify failure modes, metrics,
and a safe migration plan.
5. Architecture¶
The profile operates as a complete agent identity with bundled skills that load contextually based on the task type:
context-forge-rag profile
├── principal-ai-architect → architecture/design tasks
├── technical-roadmapping → planning/sequencing tasks
├── production-rag-architecture → retrieval design tasks
├── rag-evaluation-observability → eval/metric tasks
└── agentic-workflow-issue-factory → implementation tasks
6. CorpusIQ Use Cases¶
| Use Case | Profile Application |
|---|---|
| CorpusIQ RAG pipeline design | Design chunking strategy, embedding model selection, hybrid retrieval architecture |
| Agent evaluation framework | Define golden datasets, LLM judge rubrics, regression gates for CorpusIQ-connected agents |
| Feature roadmapping | Convert operator feedback into sequenced, dependency-mapped GitHub issues |
| Observability telemetry | Design cost tracking, trace aggregation, and performance monitoring for agent workflows |
| ADR documentation | Produce Architecture Decision Records for key technical choices |
7. Security¶
Never commit .env, API keys, OAuth tokens, session dumps, private documents, runtime memories, or customer data. The profile is designed to make architecture and implementation artifacts safer, more observable, and easier to evaluate - not to handle secrets.
8. Troubleshooting¶
| Symptom | Cause | Fix |
|---|---|---|
| Profile validation fails | Missing Python packages | pip install -r requirements.txt (check repo for requirements) |
| Skills don't load | Profile not active | Run hermes -p context-forge-rag or set as default |
| Generator smoke test fails | Template params outdated | Check templates/profile.params.yaml for required fields |
| Model errors | Provider not configured | Profile works with any Hermes-supported model provider |
← Skills Catalog | June 24 Evening Discovery →
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.