An access policy and its lists are one write, and tenants never see a Kamo service's in-cluster address
SP98 Task 1 A-7 and A-8. - A-7: AccessPolicyWriter saves the policy and replaces its lists in one transaction, retried whole on a Yugabyte conflict (@RetryOnD...
One verification run per provider across both pods
SP98 Task 1 A-6. A provider's verification (up to 200 probes on the tenant's key) holds a Redis lease, SET NX with a 10-minute TTL renewed before every probe an...
Feedback only under its own conversation, no folder cycles, logged and key-free failures, and a visitor's conversation searches the public knowledge base
SP98 Task 1 (A-1 to A-5, and the controller's knowledge-audience addition): - A-1: rating a message answers 404 unless the message belongs to the {guid} conve...
A reworded GPT-6 tool refusal is warned once, and a full effort memory forgets the oldest
Review follow-ups for the reasoning_effort none fix: - A 400/422 that names reasoning_effort and tools but is not the refusal ToolEffortFallback reads logs on...
A chat turn that offers tools requires TOOL_CALLING, and a failed probe clears supports_tool_use
Member chat and the public widget routed TEXT_GENERATION only and attached tools afterwards, so gpt-6-luna (TOOL_CALLING FAILED) took every tool-bearing - The ...
A GPT-6 tool call is asked once more at reasoning_effort none, and the model is learned
reasoning_effort "none" and refuse the default effort they apply when none is sent, so every TOOL_CALLING probe and agent step on gpt-6-luna/sol failed. - Reas...
Spend a WebSocket one-time key with one GETDEL
spendOneTimeKey read the key and then deleted it, so two handshakes racing with the same key inside its 30 s life could both read it and both open a socket for ...
A streamed member turn retries once without an optional parameter the provider refused
Member chat streams through the adapter itself, outside ProviderCallExecutor, so it had no ParameterFallback. Since SP99-T12 every Hugging Face stream asks for ...
The streamed member turn runs the member's tools and cites its sources, and a browser can open it
The /ws/ai-chat turn was text only: no Tool Plane tools, no tool loop, no knowledge-base citations, and its handshake took only the raw session id, which kamo-i...
An in-cluster model gets the time to read its prompt, inside the walk's budget
kamo-llm reads a prompt at about 110 tokens a second on k3m1 before it writes a word, so an E2 prompt past ~6k tokens spent E2's whole 60 s per-attempt bound (a...
A Hugging Face text stream is metered as the router counted it
HuggingFaceAdapter streamed through the interface's text-only default: every chunk was read for its text and nothing else, so no usage ever reached the executor...
An Ollama model's context is the window the server serves, not the trained one
OllamaDiscovery took model_info.*.context_length from /api/show: the window the model was TRAINED for. The registry listed granite4:tiny-h at 1 048 576 (and qwe...
Member chat offers the member's tools on Hugging Face models
HuggingFaceAdapter has sent tools and read tool calls since ff1f0ed, and the router routes TOOL_CALLING to it, but supportsToolUse() still answered false. ChatO...
E2's streamed usage chunk reports the metered usage, never zeros
Hugging Face, Google and Cohere text streams carry no usage event, so the assembled response counts 0 input and 0 output tokens. E2 built its stream_options.inc...
F1 takes back the auto it answers as its source language
A caller that echoes F1's from:"auto" is translated with detection instead of refused, and F2's placeholder name says why it exists. Ruling SP05-T32 (e).
Translation and OCR façades on the router for the Tool Plane
F1 POST /api/ai/v1/translations {text, to, from?} and F2 POST /api/ai/v1/ocr (multipart file, optional filename), SP05 spec §4.17: the caller's session, SP04's ...
Redeploy f9a0aa1 — run 69's deploy job hit a Kubernetes API timeout
Run 69 built f9a0aa1, but its deploy job failed at 'kubectl create ns' with 'Timeout: request did not complete within requested timeout'. No code change.
An AI member's own chat session routes, budgets and meters as actor AI
Member chat (/api/ai/sessions/**, /ws/ai-chat) routed every member as HUMAN, so an AI member's session (its SP11 workstation browser) reached UNVERIFIED models,...
Member chat uses the Tool Plane with the member's own session; MCP server admin contract fixed
Member and widget chat routed, policy-checked and metered by the router stack
Realtime relay WebSocket with single-use tickets, session cap and metering
Realtime tickets — relay tickets in Redis or OpenAI ephemeral secrets
Ephemeral secrets are minted through the gateway's shared family walk (every dialled mint metered, a rejected key marks the provider, budget alerts after the ti...
Transcription and speech façades metered in audio seconds and characters
E4 POST /api/ai/v1/audio/transcriptions and E5 POST /api/ai/v1/audio/speech, each with its internal twin, on SP01's family adapters. The WAV-header duration wal...
Embeddings on the pinned model and rerank façades
Inputs and documents reach the model in batches it accepts (TEI refuses more than its client batch of 32 with 422; a reranker's maxDocuments); malformed fields ...
POST /api/ai/agent/step with tool use, thinking replay, cache breakpoints and SSE
OpenAI-compatible chat completions with tools, images, structured output and SSE
Internal router resolve endpoint
E1 POST **************** the ordered candidates and exclusions for a capability, 424 no_capability when none, 404 org_not_found for an org that is not live (rea...
Legacy routing, /internal/text and /internal/vision on the router, with provenance
The gate's AI-member path; retire the persona outreach engine
Quiet hours convert the service clock into the contact's zone instead of relabelling it (ruling SP03-T32).
Session/internal caller context, app gate, provenance and the gateway call runtime
Candidate walk with failover, one-shot parameter fallback and metered streaming
Provider credentials never reach the log, the usage row or the rethrown error; a failed provider-health write never costs an answer; an interrupt stops the walk...
Map a Studio config onto job-title charter items
SalesAgentCharterMapper maps a Sales Agent's AgentConfig onto job-title charter items (spec 4.7.8) for the fold-in, and the six Studio templates onto the A9 per...
Compile a charter into an operating brief and a job description
RRuleText words every rule CharterRRule accepts (month-day and monthly weekday lists, an hourly rule's days) and quotes one it cannot read, so a single stored r...
Hugging Face chat sends the tools and returns the tool calls
message.tool_calls, but HuggingFaceAdapter sent no tools (so tool_choice was dropped too) and never read tool_calls. Every HUGGINGFACE model therefore failed th...
The boot backfill counts row-less-by-design models apart
Every AIService start logged 'capability rows written for 21 model(s)' though completions models no capability Kamo uses (ruling SP01-T21), so they stay row-les...
A model list anyone can read no longer passes a wrong key
The Hugging Face router and OpenRouter answer GET /v1/models with 200 for no key and for a wrong key, so Test connection read OK and sync reconciled 140 models ...
Auto routing never picks a row-less model the catalogue says Kamo cannot call
The legacy router let a model with no capability rows through on the name rule legacy-completions *-instruct) get no rows by design, and their names pass the A...
Like what you see shipping?
All of it arrives in your workspace on its own. Start on the free plan and read this page again in a month.