Back to corpus
research noteexperiment writeup candidatescore 22

Graph Kernel Service

The Graph Kernel runs as a REST API service for slice-conditioned retrieval. It's the **admissibility authority** for the semantic system.

Open full HTMLDownload HTMLCorpus JSONPaper schema

Full HTML reader

Read the full artifact

Open in new tab

Extracted abstract or opening context

The Graph Kernel runs as a REST API service for slice-conditioned retrieval. It's the **admissibility authority** for the semantic system. | Variable | Required | Default | Description | |----------|----------|---------|-------------| | `DATABASE_URL` | Yes | - | PostgreSQL connection string | | `KERNEL_HMAC_SECRET` | Yes (prod) | dev secret | HMAC secret for admissibility tokens | | `PORT` | No | 8001 | Service port | | `HOST` | No | [ip] | Bind address | | `RUST_LOG` | No | info | Log level (debug, info, warn, error) | | `LOG_FORMAT` | No | json | Log format (json, pretty) | Response includes: - `slice_id`: Unique fingerprint - `turn_ids`: Admissible turns - `admissibility_token`: HMAC-signed proof #### `POST /api/verify_token` Verify an admissibility token without HMAC secret. | Endpoint | Purpose | |----------|---------| | `GET /health` | Detailed health (database, policies) | | `GET /health/live` | Liveness probe (process alive) | | `GET /health/ready` | Readiness probe (database connected) | | `GET /health/startup` | Startup probe for Cloud Run |

Promotion decision

What has to happen next

Attach run IDs, datasets, metrics, and reproduction commands.

Why this is not always a full paper yet

Corpus pages are public-safe readers for discovered workspace artifacts. They are not automatically final papers. A corpus item becomes a polished paper only after the editable source, evidence checkpoints, references, figures, render path, and release status are attached through the paper schema.