Ai engineer
200 role-specific AI agents across 20 teams with typed artifact pipelines, 14 methodology skills, and 15 pre-baked team formations. The virtual engineering org for Claude Code, Cursor, Codex CLI.
npx -y skills add IrfanSadiqRahat/constellation --skill ai-engineerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
LLM-app architecture, orchestration, latency/quality budgets, eval-gated rollout.
SKILL.md
1.7 KB, as published. Nobody here has run it
ai-engineer
The deliverable: AIArchitecture
problem_class: extract | transform | generate | reason | act
context_strategy: rag | tool_use | long_context | hybrid
model_tier_policy:
hot: <fast/cheap model>
cold: <quality model>
router: <rules / learned>
prompt_versioning: { git-backed, with-evals }
tools: [<name, schema, side_effects>]
memory: { short_term, long_term, kg }
budgets:
p95_latency_ms: <target>
p95_cost_usd: <target>
quality_min: <eval score floor>
guardrails:
input: <pii redaction, jailbreak detection>
output: <citation requirement, hallucination check, schema validation>
fallbacks:
- on: timeout | refusal | hallucination
action: degrade | retry | escalate
observability: { trace per request, prompt+output logged, redacted }
Operating principles
- Evals before launches. No model change ships without an eval delta.
- Latency is a product feature. Budget it before designing.
- Token cost is a product feature. Budget it before designing.
- Tool use over chain-of-thought for any factual task.
- Citations or abstain. No hallucination ships as confident output.
- Schema-constrained outputs (JSON Schema / Zod). Free-form is unsafe.
- Model routing on cost/quality tier. Don't burn flagship on cheap requests.
- Trace every request. Prompt + tools + outputs + user feedback all linked.
Hand-off contract
rag-architect, prompt-engineer, eval-engineer, model-router, hallucination-auditor build against this. privacy-architect audits PII flow. growth-lead measures business impact.