The supplied case describes Novgraph as a CLI client for a hosted codebase knowledge graph, attributed to creator Shourya_55, who claims it preserves agent-written change rationales, synchronizes repository updates, and deranks stale code. It also reports claimed token savings of 78–99.74%, but the supplied web results contain no direct Novgraph coverage and establish neither its implementation nor the benchmark conditions. The snippets instead describe related tools such as CodeGraph and Graphify, supporting the broader context of persistent repository indexes for coding agents—not Novgraph’s specific capabilities or performance.
Novgraph’s claimed rationale-preserving, maintained code graph repeats positions Scott already holds in Cognitive Git and The Index Is the Data, with practical overlap in dev-wiki and search’s delta-aware indexing. No supplied radar page tracks Novgraph itself, but the unverified implementation and unexplained 78–99.74% savings establish neither an actionable improvement over Scott’s systems nor a consequential new endorsement of his position.
ip:framework.cognitive-gitip:framework.the-index-is-the-data-self-cleaning-wiki-graphdev:project.dev-wikidev:project.searchradar:cognee-codebase-memory-efficiencyradar:graphify-repository-map-contextradar:github-tool-output-cost-tradeoff
queries asked of Scott's wikis
- coding agent memory across sessions repeated repository discovery
- agent-maintained knowledge graphs change rationale decision provenance
- repository synchronization stale context invalidation retrieval ranking
- context efficiency token savings benchmarks task correctness
- hosted versus local codebase indexing agent harness integration
now 0 pts/hpeak 1 pts/hcomments 0/hpeers p26momentum: steady2 platformsage 722h
points/hour across evidence · reading as of 2026-10-12 02:59:37.977291+11:00 · deterministic, not a model opinion
2026-10-11T12:19:38Z
The attached Memgraph blog provides only category context for knowledge-graph trustworthiness; it does not validate Novgraph's specific claims. After 29 days, all Novgraph-specific evidence remains self-promotion by the creator (four HN submissions) with escalating, unexplained token-savings ranges and no independent implementation review, benchmark conditions, or task-quality results. The verification path (trynovaya.com) exists but is untested in the evidence.
2026-10-11T11:36:50Z
evidence attached: hn.story.50041665 — Memgraph blog on knowledge-graph trustworthiness directly contextualizes the hypothesis that structured repo memory displaces CLAUDE.md dumps.
2026-09-17T19:23:36Z
The latest creator post adds a trial URL and a claimed context-accounting command, but primarily repeats the product pitch with a higher, unexplained savings range. It offers a possible verification path, not evidence of a new release or demonstrated improvement in coding-agent memory.
2026-09-17T19:21:49Z
evidence attached: hn.story.49745176 — This is direct coverage of Novgraph's released persistent codebase-memory artifact and its claimed token savings.
2026-09-17T14:39:27Z
The newly attached item is another title-only HN submission by Novgraph’s creator, not an independent repository artifact as the attachment rationale suggests. It adds no implementation or validation evidence, leaving the case an unverified product claim rather than a demonstrated coding-agent memory improvement.
2026-09-17T14:22:16Z
evidence attached: hn.story.49740741 — This is an independent first-party repository artifact directly bearing on Novgraph's persistent codebase-memory hypothesis.
2026-09-13T15:30:34Z
The attached Show HN announcement suggests parallel interest in code knowledge graphs, but supplies no implementation evidence and does not independently validate Novgraph. It therefore adds category context rather than corroboration of this case’s memory-maintenance or efficiency claims.
2026-09-13T15:22:31Z
evidence attached: hn.story.49684835 — An independent released project using a knowledge graph for AI-written code corroborates the developing persistent-codebase-memory pattern.
2026-09-12T14:26:25Z
grounded: known/low — Novgraph’s claimed rationale-preserving, maintained code graph repeats positions Scott already holds in Cognitive Git and The Index Is the Data, with practical
2026-09-12T14:23:37Z
origin walked (codex/luna, conf 0.99): anchor hn.story.49672213 -> echo.github.4863c155dd by Shourya Sathwane / Novaya-AI
2026-09-12T14:22:36Z
case created — The creator describes a distinct implementation with concrete memory-maintenance mechanisms, but the supplied announcement contains neither an accessible artifact link nor supporting measurements for its savings claim.