Bhavana Sanal

What an agent can do with @bhavana-sanal's commit to melra

XAGI-Lab/melra: Open-source local MCP runtime that checks AI-agent tool calls, gates risky actions on approval, verifies results, and avoids blind retries…

• Look up how melra implements agent-runtime, agent-safety, ai-agents, with answers cited to file and line.
• Reuse a worked example: the commit “order, decisions and recovery stay straight” (3 files · TypeScript).
• Answer from melra's real TypeScript source instead of memory (AI, prompts and agents).

• Where does melra handle “order, decisions and recovery stay straight”?
• How does melra implement agent-runtime?
• What does packages/runtime-core do in melra?

Search the GitHub repository XAGI-Lab/melra with XERJ (https://hub.xerj.org/corpus/xagi-lab--melra).
Run these in a shell, in order, and check each result before the next:
1. xerj --help  (success: it prints usage). If xerj is missing: curl -fsSL https://xerj.org/get | sh
2. git clone https://github.com/XAGI-Lab/melra melra --filter=blob:none --no-checkout && git -C melra sparse-checkout set --no-cone /packages/runtime-core/src/ && git -C melra checkout ab59ffa134decd9613a378c7bd8a0e8d495c7e89
3. Start a local node and leave it running: xerj --insecure --data-dir ./xerj-data
4. xerj autoindex ./melra --url http://127.0.0.1:9200 --prefix melra --progress plain
5. xerj autoindex map --url http://127.0.0.1:9200
Then answer my question from the indexed files and cite the ax_path of every hit. Query syntax: https://xerj.org/llms.txt
Licence: AGPL-3.0. Read it for the approach only, do not copy code.

corpus xagi-lab--melra · AI, prompts and agents
Related corpora: autorag, xobotyi--cc-foundry-open-source, sandbaseai/sandbase-harness, OpenMachineLabs/openmachine
AI search and retrieval with XERJ · Case study: reference coding for agents