Replace peer-to-peer agent messaging with centralized orchestration

Before this change, agents communicated directly with each other by
writing to a shared file (Meta/agent-messages.md), following the
protocol in references/inter-agent-messaging.md. This was fragile
and hard to control.

Now the dispatcher (CLAUDE.md) is the only coordinator. Agents do
not talk to each other. Instead, when an agent detects work for
another agent (e.g. missing vault structure, orphan notes), it
outputs a "Suggested next agent" section. The dispatcher reads it,
validates it against the agents-registry, and decides whether to
chain the next agent.

New files added:
- references/agent-orchestration.md: the new coordination protocol,
  including call chain tracking, anti-recursion rules, and a max
  depth of 3 agents per user request
- references/agents-registry.md: a capability registry that maps
  each agent to its inputs, outputs, and chaining suggestions

Deleted:
- references/inter-agent-messaging.md: the old peer-to-peer protocol
This commit is contained in:
gnekt
2026-03-23 15:30:20 +01:00
parent 0a7213081c
commit 65253cdf7c
4 changed files with 155 additions and 182 deletions

View File

@@ -4,6 +4,12 @@ This reference is shared across all agents. Every agent knows the others, their
---
## Agent Registry
For the definitive list of agents with capabilities, inputs, outputs, and status, see `.claude/references/agents-registry.md`. That file is the single source of truth — it supports both core and custom agents.
---
## Language Rule
**All agents respond in the user's language.** Match the language the user writes in. If the user switches languages mid-conversation, switch with them.
@@ -109,3 +115,5 @@ All agents read `Meta/user-profile.md` for personalization. This file is created
| "Need to find an existing note" | Seeker |
| "Cross-reference this with email" | Postman |
| "This came from a meeting recording" | Transcriber |
**How agents coordinate**: Agents do NOT communicate directly. When an agent detects work for another agent, it includes a `### Suggested next agent` section in its output. The dispatcher reads this and decides whether to chain the next agent. See `.claude/references/agent-orchestration.md` for the full protocol.