Comment by Axsuul
10 hours ago
Anyone else have related reading that touches on this? I'm building my own custom harness and want to start implementing loop support, etc. But I also want to build some sort of framework so that it's dynamic (e.g. this needs to run x number of iterations, while planning needs to run y number of iterations).
OpenAI and Anthropic both have blog posts about programmatic tool use. Tbh I settled in STARLARK as the intermediary language it's a python dialect so most llms can use it with little help.https://github.com/zarldev/zarlmono/blob/main/zkit%2Fagent%2... it's a wrapper around the calls of the other tool set.
Check out https://github.com/strongdm/attractor along with the community implementations at https://factory.strongdm.ai/products/attractor#community.
Check out academic papers about:
1. Hierarchical skills, workflow, skill learning 2. Meta Harness, self-learning harnesses 3. Trace/trajectory representation 4. Common agentic benchmarks
But first more basic things like 5. Blog posts form anthropic 6. How Claude Code/PI/ Hermes!! agent works 7. Agent sessions/ Forking/ Hooks
Sounds like you want something similar to /goal mode in Codex.