consciousness/poc-memory
Kent Overstreet 3a8575b429 agents: fix vllm crash on malformed tool args, always use API
Three fixes:

1. Sanitize tool call arguments before pushing to conversation
   history — vllm re-parses them as JSON on the next request and
   crashes on invalid JSON from a previous turn. Malformed args now
   get replaced with {} and the model gets an error message telling
   it to retry with valid JSON.

2. Remove is_split special case — split goes through the normal
   job_consolidation_agent path like all other agents.

3. call_for_def always uses API when api_base_url is configured,
   regardless of tools field. Remove tools field from all .agent
   files — memory tools are always provided by the API layer.

Also adds prompt size guard (800KB max) to catch oversized prompts
before they hit the model context limit.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-20 14:33:36 -04:00
..
.claude feedback: not-relevant/not-useful commands, edge strength adjustment 2026-03-14 12:14:17 -04:00
agents agents: always use API backend, remove tools field from .agent files 2026-03-20 14:26:39 -04:00
defaults split into workspace: poc-memory and poc-daemon subcrates 2026-03-08 20:43:59 -04:00
schema transcript progress: capnp append-only log replaces stub nodes 2026-03-16 17:40:32 -04:00
src agents: fix vllm crash on malformed tool args, always use API 2026-03-20 14:33:36 -04:00
build.rs split into workspace: poc-memory and poc-daemon subcrates 2026-03-08 20:43:59 -04:00
Cargo.toml Consolidate poc-memory and poc-agent configs 2026-03-19 21:49:58 -04:00
config.example.jsonl split into workspace: poc-memory and poc-daemon subcrates 2026-03-08 20:43:59 -04:00