builder-ai
AGENTS.md2 starsChanged 5 months agoMITView on GitHub
If A Team is installed, builder-ai agents and skills slot in alongside the existing roster. No conflicts — all names are unique. builder-ai agents can be dispatched by the A Team orchestrator like any other specialist.
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,700 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ai-cost-auditSkill
- ai-safety-reviewSkill
- context-optimizationSkill
- eval-before-shipSkill
- fallback-requiredSkill
- model-benchmarkingSkill
- prompt-versioningSkill
- rag-pipeline-designSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 33
If A Team is installed, builder-ai agents and skills slot in alongside the existing roster.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
builder-ai — Agent & Skill Roster v1.0.0
Hard Gate Skills
| Skill | Description | Triggers |
eval-before-ship | Blocks LLM features without documented eval results | Before any PR that adds or modifies a prompt or model |
prompt-versioning | Enforces prompt version control in prompts/ | Whenever writing or modifying a prompt |
fallback-required | Blocks LLM calls without defined fallback paths | Before any PR that adds an LLM API call |
Workflow Skills
| Skill | Description | Triggers |
rag-pipeline-design | RAG pipeline design: chunking, embedding, retrieval, reranking, generation | Designing or auditing a RAG feature |
model-benchmarking | Structured model comparison with cost-quality analysis | Selecting a model for any feature |
context-optimization | Context reduction: prompt trimming, chunk reduction, caching | Latency or cost issues, or approaching context limits |
ai-cost-audit | LLM spend audit and projection | Pre-launch cost estimate or unexpected cost growth |
ai-safety-review | Safety review: injection, hallucination, abuse, agentic scope | Before any user-facing LLM feature ships |
Agents
| Agent | Tier | Model | Trigger |
prompt-engineer | Tier 2 | claude-sonnet-4-6 | Writing, iterating, or debugging prompts |
eval-designer | Tier 2 | claude-sonnet-4-6 | Designing eval suites, writing harness code |
rag-architect | Tier 1 | claude-opus-4-8 | Designing or debugging retrieval pipelines |
model-selector | Tier 2 | claude-sonnet-4-6 | Model comparison and selection |
ai-safety-reviewer | Tier 1 | claude-opus-4-8 | Safety review before user-facing features ship |
Integration with A Team
If A Team is installed, builder-ai agents and skills slot in alongside the existing roster. No conflicts — all names are unique. builder-ai agents can be dispatched by the A Team orchestrator like any other specialist.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.