Skip to content
Elva by Theneo
Sign up free
BLOG

Notes for the agentic era

Engineering deep-dives, product decisions, and what we learn shipping APIs to agents.

Testing · 7 min read

API testing tools that generate tests from code, compared

Spec-driven fuzzers, traffic-recording tools, AI request assistants, and security-focused generators all claim to "generate API tests." A researched map of what each actually generates, from what source, and what it misses.

Read the post →
API Documentation
Automated API documentation without an OpenAPI specMost API doc platforms assume you already have an OpenAPI file. A researched map of what actually produces a reference doc when you don't: collection imports, AI write-ups, traffic discovery, and code-derived catalogs.7 min read
Governance
API changelog tools, comparedSpec-diff generators, docs platforms with a changelog built in, generic release-notes widgets, and git-native automation: a researched map of what actually produces an API changelog, and where each one falls short.8 min read
MCP
MCP logging and observability tools, comparedA researched map of how teams actually watch MCP traffic today: the protocol's own shift to OpenTelemetry, the observability platforms that added MCP tracing, and the gap between debugging a server and running one.8 min read
MCP
Tools to build governed MCP servers for agents, comparedA researched map of what actually governs an enterprise MCP server: SDKs that write server code, gateways that add auth and rate limits in front of one, and platforms that generate and host a governed server directly.9 min read
MCP
Turning a code repository into an MCP server, comparedWhat actually turns a codebase into an MCP server: GitHub's own API server (a different job entirely), OpenAPI-to-MCP generators that need a spec first, code annotation frameworks, and the thin no-spec category.8 min read
Governance
How teams actually control what AI agents can accessA researched map of the real tools behind "agent permissions": policy engines, non-human identity platforms, MCP's own OAuth flow, and cloud IAM for agents — four different layers people keep treating as one.8 min read
AI Agents
API discovery platforms and marketplaces for AI agents in 2026A researched map of how AI agents and the teams building them find and call APIs at scale outside of MCP: public marketplaces, unified-API platforms, and OpenAPI-driven agent tool generation, and where each one stops short of governing who gets to call what.8 min read
Contracts
How teams maintain API contracts without a manual specA precise map of the tools people reach for to keep an API contract true without hand-writing one: schema-first frameworks, code-reading generators, contract testing, and drift detection, and where each one actually stops.8 min read
MCP
The best MCP tool discovery platforms in 2026A researched comparison of where AI agents and developers actually find MCP servers today — the official registry, community marketplaces, integration platforms, and packaged hosting — plus where governed private serving fits.8 min read
Governance
An API governance checklist that fits in one pipelineTwelve checks that make API governance real: catalog, contracts, breaking change gates, auth, observability, and agent access, all enforced in CI.4 min read
Governance
API versioning strategies that survive breaking changesURL versions, headers, dates, or no versions at all: how the four common strategies handle a breaking change, and the discipline that matters more.5 min read
OpenAPI
How to generate an OpenAPI spec from existing codeFour working ways to get an OpenAPI 3.1 spec out of a codebase that never had one, what each approach misses, and how to keep the result true.4 min read
MCP
MCP vs REST: what changes when agents call your APIMCP does not replace REST. It changes who the caller is. What that shift means for interface design, auth, error messages, and operations.4 min read
MCP
What is an MCP server, explained for API teamsA plain explanation of MCP servers: what the protocol does, how a server maps to your API, and what it takes to run one in production.5 min read
MCP
MCP servers are infrastructure. Run them like it.Weekend demos die on Monday. What we learned running MCP servers for real agent traffic, and what changes when a tool call has an SLA.8 min read
Engineering
How Elva reads a repo: from route handlers to OpenAPI 3.1Static analysis, framework fingerprints, and the messy middle where code and spec disagree.11 min read
Product
July: field-level constraints, key-scoped analytics, drift alertsEverything that shipped this month, including the schema editor and per-key usage.3 min read
Product
Contracts: one API, four audiencesPartners, internal teams, the public, and agents all need different views of the same endpoints. Contracts make that explicit.6 min read
Engineering
Scoring agent readiness: what the A–E grade actually measuresDescription coverage, schema completeness, auth clarity, and the weights behind the grade.9 min read
MCP
23 tools, 4 agents, zero surprisesPer-agent activity, tool-level latency, and catching the schema gap before Claude does.7 min read
Product
Why “no spec” can’t mean “no catalog”Half the APIs that matter have no OpenAPI file. The catalog has to start from code, not from paperwork.5 min read
MCP
What 1.2 million agent tool calls taught us about timeouts, retries, and the true cost of a vague error messageNinety days of production MCP traffic in numbers: retry storms, timeout cliffs, and why one ambiguous error string cost more than any outage did.1 min read

BROWSE BY TOPIC

MCPGovernanceAI AgentsOpenAPIContractsEngineeringObservabilityProductInfrastructure

Ship notes, monthly

One email with what shipped and what we learned. Unsubscribe anytime.