docs(track3): submission summary, architecture diagram, and demo script#351
Open
kaveone wants to merge 15 commits into
Open
docs(track3): submission summary, architecture diagram, and demo script#351kaveone wants to merge 15 commits into
kaveone wants to merge 15 commits into
Conversation
- TRACK3_SUBMISSION.md: mermaid architecture + identity-mesh diagrams, mandate-by-mandate mapping, r19 workflow table, live proof (run 57b1b6b8, PR #4 merged, issue #3 closed), service endpoints. - DEMO_SCRIPT.md: ~4-5min screen-recording timeline that runs the same cloud-native pipeline on a fresh issue, with narration cues, an auto-approve helper, and reset/gotcha notes. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
✅ Deploy Preview for lustrous-pasca-e9664a ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
Second-level final-cut timeline (8 beats), continuous narration, auto-approved gates, and a time-lapsed coder segment so a ~6-min run fits 2 minutes. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…GitHub) Rework the 2:00 script around three windows — dashboard [DASH] as the main product view, Cloud Run monitor [LOGS] proving on-GCP execution, GitHub [GH] showing issue->PR->merge->close. Add scripts/demo-logs.sh: colored, service-tagged Cloud Run log tail (real-time alpha tail, 3s-poll fallback) for the logs window, plus a Logs Explorer streaming query for the console. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Lower-third caption track synced to the 2:00 beats (mandate keywords readable sound-off), persistent window corner-tags, and a paste-ready pinned video description. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The demo repo's revka-issue-trigger.yml Action fires on the 'revka-demo' label and POSTs to the Cloud Run orchestrator (bearer token in repo secrets) — restore that as the headline flow (issue -> Action -> Cloud Run), with the manual curl POST kept as a fallback. REVKA_GATEWAY_URL repointed from the old tunnel to the Cloud Run URL. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- docs/ENTERPRISE_ROADMAP.md: GKE hybrid topology (stateful orchestrator + Filestore-backed durable artifacts; managed agent tier), the native-SDK API-key hard rule (no consumer subscriptions in the enterprise tier), the continuous goal-driven autonomous-improvement product vision (Kumiho as the grounding flywheel), a hardening backlog, and mandate alignment. - TRACK3_SUBMISSION.md: automatic revka-label trigger flow, the grounded reviewer citing 'Rule 1' from Vertex AI Search, refreshed run/PR/issue evidence (#8/#9, #6/#7), and a note that the cloud workflow is a deployment-owned Kumiho artifact, not an OSS builtin. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…o enterprise diagram - Submission diagram: triggers are now pluggable (GitHub label = one demo example, not the architecture core); centers Revka on GCP and its service integrations (Vertex, Vertex AI Search, Secret Manager, Artifact Registry, A2A agents, control plane), with the GitHub repo as the work target and Kumiho shown as external memory. - Enterprise diagram: add the BYO / cost-sensitive agent tier (subscription CLIs / customer keys, customer-owned ToS) alongside the API-key enterprise tier. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A clean, read-aloud narration script (~300 words, ~150 wpm) synced to the 8 cut points, with pause markers and pacing notes (what to trim/hold to hit 2:00). Complements the per-beat VO cues and the lower-third captions. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Revka is an autonomous, auditable, visually-defined workflow PLATFORM; the github-issue-resolver is one example workflow, not the product. Rewrite the voice-over to open and close on the platform (visual workflows, any multi-step enterprise process — software, security, data, research, ops — governed and audit-trailed) and frame the issue-resolution run as one example in the middle. Align the pinned description and opening captions to the same platform-first framing. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…ntiation→proof) Add a full presenter script that argues the case rather than narrating the screen: the enterprise problem (autonomous agents are black boxes — no oversight, no memory, no audit trail — so they can't ship), how Revka solves it (visual governed workflows, human gates, on Google Cloud), and what makes it different from other agents — graph-native memory that compounds across runs, plus infrastructure that records every workflow output as an auditable trail. The demo becomes evidence in the middle, not the subject. Keep the tight 2:00 voice-over as a compressed alternative. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…umiho RAG Reframe the submission around the actual rubric: - Platform-first intro; github-issue-resolver is the demonstrated example. - New 'five key considerations' mapping table (ADK orchestration, Agent Engine deployment, business use case, grounding/RAG, multi-agent > single). - Hybrid runtime: reviewer on Vertex AI Agent Engine (reasoning + Vertex AI Search), coder on Cloud Run (git/shell sandbox) — 'right runtime per agent'. - Kumiho framed as graph-native memory / private-data RAG that stores and versions every workflow run — the heart of Revka. - Explicit 'why multi-agent + grounding beats a single agent' argument. - Demo narration + JUDGES updated to name Agent Engine and Kumiho. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Reviewer is live on Vertex AI Agent Engine (reasoningEngines/3625053003137941504), verified returning a grounded verdict citing Rule 1 (money is integer cents). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two documents for the Google for Startups AI Agents Challenge (Track 3):
docs/TRACK3_SUBMISSION.md— mermaid architecture + identity-mesh diagrams, mandate-by-mandate mapping (cloud-native runtime, Gemini/Vertex, A2A, B2B + ADK), the r19 8-step workflow table, and live proof: run57b1b6b8, PR Bump hmac from 0.12.1 to 0.13.0 #4 merged, issue Bump rustls from 0.23.38 to 0.23.39 in the rust-all group #3 closed on the demo repo.docs/DEMO_SCRIPT.md— a ~4-5 minute screen-recording timeline that runs the same cloud-native pipeline on a fresh issue, with narration cues per segment, an auto-approve helper for hands-free gates, and reset/gotcha notes.🤖 Generated with Claude Code