architect: agent management redesign #21

Merged

disinto-admin merged 1 commit from architect/agent-management-redesign into main

2026-04-15 17:38:55 +00:00

architect-bot commented

2026-04-12 02:03:19 +00:00

Collaborator

Sprint pitch: agent management redesign

Vision issue: #557

What this enables

Operators can hire agents by backend (disinto hire anthropic, disinto hire llama --url ...) instead of inventing names and roles, and list all agents (disinto agents list) with backend, model, roles, and status in one table.

The factory becomes self-describing: an operator who inherits a running instance can immediately see what agents exist, what backends they use, and what roles they fill.

Complexity

~4 files touched (bin/disinto, lib/hire-agent.sh, lib/generators.sh, docker/agents/entrypoint.sh)
4-5 sub-issues estimated
~80% gluecode refactoring, ~20% greenfield

Risks

Old hire-an-agent must keep working during transition (deprecation path)
6 core agents hardcoded in multiple places — making dynamic needs careful testing
TOML schema must remain backward-compatible

Cost

No new services, cron jobs, or formulas. Pure CLI/config refactor. ~230 lines new, some removed.

Recommendation

Worth it. High-value, low-risk adoption improvement. The current UX requires knowing name, role, and backend in the wrong order. Independent of all in-flight sprints.

Reply ACCEPT to proceed with design questions, or REJECT: <reason> to decline.

## Sprint pitch: agent management redesign **Vision issue:** #557 ### What this enables Operators can hire agents by backend (`disinto hire anthropic`, `disinto hire llama --url ...`) instead of inventing names and roles, and list all agents (`disinto agents list`) with backend, model, roles, and status in one table. The factory becomes self-describing: an operator who inherits a running instance can immediately see what agents exist, what backends they use, and what roles they fill. ### Complexity - ~4 files touched (bin/disinto, lib/hire-agent.sh, lib/generators.sh, docker/agents/entrypoint.sh) - 4-5 sub-issues estimated - ~80% gluecode refactoring, ~20% greenfield ### Risks - Old `hire-an-agent` must keep working during transition (deprecation path) - 6 core agents hardcoded in multiple places — making dynamic needs careful testing - TOML schema must remain backward-compatible ### Cost No new services, cron jobs, or formulas. Pure CLI/config refactor. ~230 lines new, some removed. ### Recommendation **Worth it.** High-value, low-risk adoption improvement. The current UX requires knowing name, role, and backend in the wrong order. Independent of all in-flight sprints. --- Reply `ACCEPT` to proceed with design questions, or `REJECT: <reason>` to decline.

architect-bot added 1 commit 2026-04-12 02:03:19 +00:00