From dbef8d6244e9633d2c8d607f51ef69e6cc36a374 Mon Sep 17 00:00:00 2001 From: jenkins Date: Thu, 6 Aug 2026 05:15:37 -0300 Subject: [PATCH] docs(runbook): Claude primary, hung-build escalation, narrowed alerting Co-Authored-By: Claude Opus 5 --- knowledge/runbooks/hermes-triage-demo.md | 21 ++++++++++++++++--- .../comms/knowledge/hermes-triage-demo.md | 21 ++++++++++++++++--- 2 files changed, 36 insertions(+), 6 deletions(-) diff --git a/knowledge/runbooks/hermes-triage-demo.md b/knowledge/runbooks/hermes-triage-demo.md index fa9469fa9..99a1d8cbd 100644 --- a/knowledge/runbooks/hermes-triage-demo.md +++ b/knowledge/runbooks/hermes-triage-demo.md @@ -15,13 +15,28 @@ You arm the failure (SEED_FAILURE=true) -> Ariadne calls the Hermes Agent API (/v1/runs) with $triage-titan-test-failures -> Hermes returns schema-valid diagnosis + requested_action - -> Ariadne authorizes (nine gates) and, if remediation is enabled, - creates Job hermes-demo-repair- (writes "healthy") + -> Ariadne authorizes (twelve gates) and, if remediation is enabled, + patches the fixture ConfigMap back to "healthy" in process -> Ariadne triggers ONE rebuild with SEED_FAILURE=false -> rebuild passes -> incident resolved -Anything else -> human_required metric -> vmalert HermesTriageHumanRequired +Anything else -> issue in the service repo + human_required metric ``` +The model behind Hermes is `anthropic/claude-opus-5` as of 2026-08-06, with +`openai-codex/gpt-5.6-terra` as first fallback and a local `gpt-oss:20b` +behind that, so an expired Anthropic credential degrades rather than stops. + +Two behaviours worth knowing before you demo, because both look like nothing +happening: + +- **A build that never finishes** is escalated once it passes + `ARIADNE_HERMES_HUNG_BUILD_MINUTES` (default 45). No model is consulted - + the console is still being written - so the issue says only that the build + overran and is holding an agent slot. +- **Alerting no longer fires on every escalation.** The issue in the service + repository is the durable artifact. Email now means either a repair ran and + failed, or an escalation has sat untouched for six hours. + ## Arming the demo Jenkins UI: `https://ci.bstein.dev/job/hermes-triage-demo/` → *Build with diff --git a/services/comms/knowledge/hermes-triage-demo.md b/services/comms/knowledge/hermes-triage-demo.md index fa9469fa9..99a1d8cbd 100644 --- a/services/comms/knowledge/hermes-triage-demo.md +++ b/services/comms/knowledge/hermes-triage-demo.md @@ -15,13 +15,28 @@ You arm the failure (SEED_FAILURE=true) -> Ariadne calls the Hermes Agent API (/v1/runs) with $triage-titan-test-failures -> Hermes returns schema-valid diagnosis + requested_action - -> Ariadne authorizes (nine gates) and, if remediation is enabled, - creates Job hermes-demo-repair- (writes "healthy") + -> Ariadne authorizes (twelve gates) and, if remediation is enabled, + patches the fixture ConfigMap back to "healthy" in process -> Ariadne triggers ONE rebuild with SEED_FAILURE=false -> rebuild passes -> incident resolved -Anything else -> human_required metric -> vmalert HermesTriageHumanRequired +Anything else -> issue in the service repo + human_required metric ``` +The model behind Hermes is `anthropic/claude-opus-5` as of 2026-08-06, with +`openai-codex/gpt-5.6-terra` as first fallback and a local `gpt-oss:20b` +behind that, so an expired Anthropic credential degrades rather than stops. + +Two behaviours worth knowing before you demo, because both look like nothing +happening: + +- **A build that never finishes** is escalated once it passes + `ARIADNE_HERMES_HUNG_BUILD_MINUTES` (default 45). No model is consulted - + the console is still being written - so the issue says only that the build + overran and is holding an agent slot. +- **Alerting no longer fires on every escalation.** The issue in the service + repository is the durable artifact. Email now means either a repair ran and + failed, or an escalation has sat untouched for six hours. + ## Arming the demo Jenkins UI: `https://ci.bstein.dev/job/hermes-triage-demo/` → *Build with