https://enola.tech/blog/ MCP servers, AI coding agents, and software architecture from the enola team. en Wed, 14 Oct 2026 09:00:00 GMT https://enola.tech/blog/dogfooding-enola-typescript-extractor https://enola.tech/blog/dogfooding-enola-typescript-extractor How a missed N+1 led to better TypeScript detection, lower extractor complexity, and 1.3–3.6× faster extraction with byte-identical output. Wed, 14 Oct 2026 09:00:00 GMT Gert https://enola.tech/blog/what-enola-saves-in-real-agent-sessions https://enola.tech/blog/what-enola-saves-in-real-agent-sessions Across 238 Claude Code and OpenCode sessions, 4,375 Enola calls added ~4.1M tokens to context. See latency, tool costs, and source-reading estimates. Sun, 11 Oct 2026 09:00:00 GMT Dejan Menges benchmarks tokens coding agents MCP Claude Code OpenCode https://enola.tech/blog/does-your-repository-already-have-rules https://enola.tech/blog/does-your-repository-already-have-rules An introduction to Enola’s constraint miner: find dependency patterns, review exceptions with your team, and turn agreed boundaries into checks. Gert https://enola.tech/blog/what-makes-a-coding-agent-correct-its-own-mistakes https://enola.tech/blog/what-makes-a-coding-agent-correct-its-own-mistakes Validator feedback changes whether a coding agent can repair rejected work. Location, observed state, and valid alternatives make the next attempt better. Tue, 29 Sep 2026 09:00:00 GMT Gert coding agents agent harnesses validator feedback software architecture https://enola.tech/blog/package-metrics-on-enola https://enola.tech/blog/package-metrics-on-enola We ran enola's package metrics on its own 117 typed packages. Nine were flagged; one warranted structural work, and the useful changes produced byte-identical facts across 91 repositories. Wed, 23 Sep 2026 09:00:00 GMT Dejan package metrics software architecture refactoring static analysis coding agents https://enola.tech/blog/dead-code-is-a-context-tax https://enola.tech/blog/dead-code-is-a-context-tax A graph can tell you nothing references a symbol. Whether to delete it is a separate decision, and it needs context the graph does not hold. How we cleared forty symbols out of Enola with zero regressions. Sat, 19 Sep 2026 09:00:00 GMT Dejan dead code technical debt static analysis coding agents context window https://enola.tech/blog/the-harness-decides-what-the-model-sees https://enola.tech/blog/the-harness-decides-what-the-model-sees A coding model never sees the whole repository or session. The harness selects its context, tools, persistent state, and feedback at every step. Thu, 17 Sep 2026 09:00:00 GMT Gert https://enola.tech/blog/how-to-use-enola-dashboard https://enola.tech/blog/how-to-use-enola-dashboard Open an architecture snapshot, investigate findings and dependencies, compare a change with a baseline, and verify snapshot freshness and coverage. Tue, 15 Sep 2026 09:00:00 GMT Gert https://enola.tech/blog/instructions-are-not-controls https://enola.tech/blog/instructions-are-not-controls Repository instructions guide coding agents before a change. Deterministic checks establish whether the result complied—and gates decide what happens next. Sun, 13 Sep 2026 09:00:00 GMT Gert https://enola.tech/blog/policy-as-code https://enola.tech/blog/policy-as-code A worked example of PCI DSS and GDPR-inspired constraints that check code reachability and policy coverage across a Go module. Tue, 08 Sep 2026 09:00:00 GMT Dejan policy as code compliance PCI DSS GDPR architecture constraints CI SARIF https://enola.tech/blog/how-we-benchmark-a-code-graph https://enola.tech/blog/how-we-benchmark-a-code-graph Eight benchmark scripts cover reproducibility, regression precision, memory, architecture labels, command contracts, agent behaviour, and changes between sweeps. Sat, 05 Sep 2026 09:00:00 GMT Dejan benchmarks static analysis reproducibility regression testing coding agents methodology https://enola.tech/blog/technical-debt-additional-cost https://enola.tech/blog/technical-debt-additional-cost A structural condition, like a service reading another team's database directly, isn't technical debt by default. It becomes debt only against a counterfactual: what a reasonable alternative would save on the work ahead. What that means for pull requests, baselines, and CI policy. Wed, 02 Sep 2026 09:00:00 GMT Gert https://enola.tech/blog/testing-coding-agent-repositories https://enola.tech/blog/testing-coding-agent-repositories Enola found a wide slash-command dispatcher. Tracing the metric revealed 89 call targets, 97 constructed targets, and one declaration edge. Tue, 25 Aug 2026 09:00:00 GMT Gert architecture audit OpenAI Codex coding agents fan-in fan-out static analysis https://enola.tech/blog/what-is-an-agent-harness https://enola.tech/blog/what-is-an-agent-harness A model generates decisions; the harness around it makes those decisions operational: tools, context, execution, permissions, and feedback. Where the term ends, how it differs from a framework or a prompt, and where enola fits in the loop. Sat, 22 Aug 2026 09:00:00 GMT Gert https://enola.tech/blog/what-a-code-graph-costs-in-memory https://enola.tech/blog/what-a-code-graph-costs-in-memory About 3 KB per fact while generating, about 0.7 KB per fact once served, and a 5.7x gap between the two. What Enola costs in RAM from a small service to the Linux kernel, why it is written in Go, and what supporting twenty languages does to the bill. Thu, 20 Aug 2026 09:00:00 GMT Dejan memory Go performance architecture graph static analysis benchmarks https://enola.tech/blog/how-to-use-enola-day-to-day https://enola.tech/blog/how-to-use-enola-day-to-day Turn the session hook off when you want speed and keep the gate in CI, read the --explain report, write a failure policy that fits your language, manage baselines, and read enola log — with real command output from Gitea, Tokio, Excalidraw and a demo service. Tue, 18 Aug 2026 09:00:00 GMT Dejan Menges https://enola.tech/blog/what-is-a-dependency-graph https://enola.tech/blog/what-is-a-dependency-graph A dependency graph records one relation from specific evidence, not a universal truth about your code. What nodes and edges actually mean, why cycles aren't always a defect, and what reverse traversal proves versus what it doesn't. Sat, 15 Aug 2026 09:00:00 GMT Gert https://enola.tech/blog/architecture-fitness-sensor https://enola.tech/blog/architecture-fitness-sensor Harness engineering splits the controls around a coding agent into guides and sensors, computational and inferential. Mapping Enola onto that grid, including the three quarters of it Enola deliberately does not fill. Mon, 10 Aug 2026 09:00:00 GMT Dejan Menges https://enola.tech/blog/how-did-the-codebase-get-like-this https://enola.tech/blog/how-did-the-codebase-get-like-this Enola can now record and explore how a codebase's architecture changed over time. enola log, show, diff, and blame turn architectural drift into evidence you can query, instead of a story you reconstruct by hand from old commits. Tue, 04 Aug 2026 09:00:00 GMT Dejan architectural history enola log structural blame regression testing AI coding agents static analysis https://enola.tech/blog/cross-repository-code-analysis https://enola.tech/blog/cross-repository-code-analysis Relationships between repositories are rarely written down in one place. Why cross-repository dependency analysis is an evidence problem, not a search problem. Tue, 28 Jul 2026 09:00:00 GMT Dejan cross-repository analysis multi-repo architecture dependency graph AI coding agents static analysis MCP https://enola.tech/blog/what-is-a-token-llm https://enola.tech/blog/what-is-a-token-llm Tokens are how every model reads and generates text. What they are, how byte pair encoding splits words into pieces, and why some languages cost several times more than English. Mon, 27 Jul 2026 09:00:00 GMT Gert tokenization byte pair encoding LLM pricing context windows multilingual AI token economics https://enola.tech/blog/enola-chatui-sveltekit-benchmark https://enola.tech/blog/enola-chatui-sveltekit-benchmark Coupling hotspots, two cyclic dependencies wrapped around the app's most complex function, and a handful of auth-related god classes. Sun, 19 Jul 2026 09:00:00 GMT Gert architecture audit SvelteKit chat-ui impact analysis cyclic dependencies god classes https://enola.tech/blog/parsing-code-is-not-mapping-architecture https://enola.tech/blog/parsing-code-is-not-mapping-architecture A syntax tree is not an architecture graph. A symbol index is not one either. They can be inputs into one, but they are not the thing itself — and that gap is where most of the real engineering work lives. Fri, 17 Jul 2026 09:00:00 GMT Dejan architecture Tree-sitter symbol graph multi-repo MCP coding agents deterministic analysis https://enola.tech/blog/four-memory-layers-coding-agent https://enola.tech/blog/four-memory-layers-coding-agent Most coding agents have one memory layer: whatever is in the context window right now. The full picture is four layers — and most teams are missing at least two. Wed, 15 Jul 2026 09:00:00 GMT Gert agent memory context engineering structural memory long-term memory coding agents MCP https://enola.tech/blog/token-economics-coding-agent https://enola.tech/blog/token-economics-coding-agent Input, output, cache writes, cache hits: where token costs actually land in an agentic coding loop, and what it's worth to close the context gap before the agent's first tool call. Thu, 09 Jul 2026 09:00:00 GMT Gert token economics LLM pricing context gap prompt caching coding agents MCP https://enola.tech/blog/the-most-boring-file-in-enola https://enola.tech/blog/the-most-boring-file-in-enola The extractor cache's version history isn't glamorous, but its comments read like a changelog of every place a naive parser gets architecture wrong. Tue, 07 Jul 2026 09:00:00 GMT Dejan static analysis extractor cache architecture graph deterministic analysis coding agents MCP https://enola.tech/blog/claude-code-context-loop https://enola.tech/blog/claude-code-context-loop Gather context, take action, verify. Of those three phases, "gather context" has no fixed procedure — it's done by searching and reading, building an approximation in every fresh session. That's the phase a deterministic graph changes. Fri, 03 Jul 2026 09:00:00 GMT Gert Claude Code agentic loop context window MCP architecture dependency graph deterministic analysis https://enola.tech/blog/enola-apache-airflow-benchmark https://enola.tech/blog/enola-apache-airflow-benchmark Coupling hotspots, cyclic dependencies, god classes, complexity outliers — and a 56,308-node blast radius hiding inside one module. We walked through the exact review a staff engineer would do. Sun, 28 Jun 2026 09:00:00 GMT Gert architecture audit Apache Airflow impact analysis cyclic dependencies god classes blast radius pre-refactor review https://enola.tech/blog/codebase-memory-is-not-an-architecture-map https://enola.tech/blog/codebase-memory-is-not-an-architecture-map Most codebase-memory tools parse symbols and retrieve fragments. That is useful, but it is not the same as building an architectural model. The difference matters. Thu, 25 Jun 2026 09:00:00 GMT Dejan architecture codebase memory knowledge graph MCP coding agents RAG deterministic analysis https://enola.tech/blog/onboard-new-joiners-faster https://enola.tech/blog/onboard-new-joiners-faster The true cost of onboarding isn't the paperwork — it's the three weeks a new engineer spends building a mental model of the codebase, and the three weeks of everyone else's time it drains. Sun, 21 Jun 2026 09:00:00 GMT Gert onboarding developer experience MCP codebase architecture knowledge graph Claude Code AI coding https://enola.tech/blog/what-are-explore-agents https://enola.tech/blog/what-are-explore-agents The hidden cost of how AI coding tools understand your codebase. Every AI coding tool is an explore agent — and the exploration has a measurable cost in tokens, time, and money. Fri, 19 Jun 2026 09:00:00 GMT Gert explore agents MCP AI coding token costs Claude Code Cursor codebase architecture LLM benchmarks