6 Commits

Author	SHA1	Message	Date
alvis	4e67186c08	docs: A2A-3 — task schema, field-by-field KB-literal mapping ... Adds §6a documenting how each Task field (id, intent, required tier, target queue, priority, status, context refs, result ref, submitter, deadline) is actually represented in Kanboard today (columns, tags, score, priority, comments, native fields), per the already-agreed KB-literal decision (§10.2). Ground truth taken from kb-claim and kb_worker.py, not written in the abstract. Kanboard #135	2026-07-22 11:06:34 +00:00
alvis	df2071d5ee	docs: v2.1 — no metered API by default; embedding-based routing ... alvis: the workflow is scoped to Claude Code + Kimi wrapper + local GPU embedder + small local model. We do not pay per-token for API usage. - NEW §3a cost model: the binding constraints are QUOTA and VRAM, not money. The "cheap tier" is the LOCAL small model, not a cheap paid one; paid deployments are opt-in fallback per virtual key and never route implicitly. Claude Code is a flat-subscription runtime, not a metered API. - Routing classification is EMBEDDING-based on the already-resident local bge-m3 (LiteLLM Auto Router / semantic-router takes a configurable embedding_model) — no classifier LLM, no API spend. Keep a heuristic fallback: Auto Router v2 is a week old with an open embedding bug report. - Budget governor reframed: arbitrates quota and GPU, not spend. - GPU residency section renumbered 3b. - Decision log 23-25 (incl. Hindsight caller-supplied cognition mode as an optional extra path, default API/queue path unchanged). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014t8Qg9gi7H7HtT8MncoXAB	2026-07-21 09:02:24 +00:00
alvis	1db3dd1b34	docs: v2.1 fix — cron catch-up leaves the keeper; faults are incidents ... alvis: a missed cron is an operational fault, not a task-lifecycle event, and re-firing it would make the keeper a trigger. Removed cron catch-up from fabric-keeper duties entirely. - Keeper is now strictly janitorial: neither assigns NOR triggers work (lease sweeps, deadlines, dead-letter, digests, metrics only). - Schedules that must not be missed are monitored in Zabbix like any other infra fault. - NEW §6d: faults are incidents, handled PER PROBLEM CLASS by a future ops-agent (watches Zabbix, applies known mitigation or files a task with the incident as context) — never a blanket catch-up policy, since a missed briefing / backup / consolidation want different responses. - Decision log entry 22. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014t8Qg9gi7H7HtT8MncoXAB	2026-07-21 08:03:27 +00:00
alvis	714a9ca785	docs: agent platform v2.1 — hardening review (agreed 2026-07-21) ... Fixes two v2 design bugs and closes the review gaps alvis accepted: - Transport rule: sync = direct A2A RPC (journaled to KB); async = KB polling; polling never on a sync path. - GPU residency: local a(t)=f(VRAM), never-evict set, pre-load checks. - NEW §5b Humans (plural): per-human private banks + shared household bank, interlocutor-scoped recall, cross-human leakage forbidden; human inbox design (priority-routed notifications, digest, no default-approve, vacation mode). - NEW §6b fabric-keeper: scheduling decomposed (triggering/dispatch/ admission/time); the keeper janitor owns leases, deadlines, cron catch-up, dead-letter, digests. - NEW §6c KB as tier-0 (backup/monitoring/degraded mode). - Completion vs verification: Done column = unverified, closed task = verified, producer never closes own task. - A2A auth mandatory (LAN untrusted: VPN peers); fabric hygiene (creation quotas, ancestry cap, dead-letter); personas/Cards in git; golden-set eval gate on backbone swaps. - Decision log entries 12-21. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014t8Qg9gi7H7HtT8MncoXAB	2026-07-21 07:34:19 +00:00
alvis	7be30c7174	docs: agent platform design v2 — the agent algebra (agreed 2026-07-21) ... Unified redesign: two axioms (everything that can receive work is an Agent with a Card and availability function; one verb submit(task, target)) from which the v1 rules follow as theorems — per-model queues (scarcity), quota parking (a(t)=0), reflect-as-task, backbone swap as constraint edit, the Claude Code loop as an ordinary consumer, human-as-agent (KB waiting column = his inbox). Adds what v1 lacked: trust classes (vault=trusted only), prompt-injection taint boundary, always-ask escalation, global budget governor, per-task workspace-lease sandboxing with PR-only merges, KB-literal fabric decision, LiteLLM Auto Router v2 for sync routing, thin KB-polling workers as the executor, Langfuse kept as observability, real A2A protocol adoption. Includes decision log from alvis. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014t8Qg9gi7H7HtT8MncoXAB	2026-07-21 07:13:54 +00:00
alvis	2499218c18	docs: add Agap agent platform design (A2A, per-model queues, agents) ... Overall design for moving from a single tool-heavy Adolf to a multi-agent platform: agents as personas (prompt+memory+tools) that can swap backbones, models as the scarce queued resource (per-model queues with always-on / quota-gated / cost-gated / on-demand lifecycles), and A2A patterns (agent cards, task lifecycle, context-by-reference) to move work between them. Covers the required cases: Hindsight reflect as an async queued task to the large model, the Claude Code loop as just another queue-consuming agent, why queues are keyed by model rather than agent, and how the lab grows to remote llama / extra GPUs / research agents. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014t8Qg9gi7H7HtT8MncoXAB	2026-07-21 06:03:29 +00:00