infer-nfr-spec-from-code

Infers enforced non-functional requirement targets from scan-index.json during codify workflows.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill infer-nfr-spec-from-code
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: infer-nfr-spec-from-code
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/skills/infer-nfr-spec-from-code
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill infer-nfr-spec-from-code

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents teams from guessing non-functional requirements by reconstructing what the shipped system already enforces, using evidence from scan-index outputs instead of inventing targets.

Core Features & Use Cases

  • Grounded NFR Reconstruction: Builds a per-NFR contract of enforced targets by tracing signals (idioms, manifests, deploy config, CI) back to concrete evidence paths.
  • Mechanism + Verification Pairing: For every inferred NFR entry, captures both the delivery mechanism (library, middleware, layer, deploy config block) and a verification method (test/CI/monitor signal).
  • No-Invention Guardrails: Records missing or unsupported signals as knowledge gaps rather than fabricating thresholds or verification steps.
  • Codify Pipeline Integration: Runs during /codify after quality and physical architecture/quality-profile proposals exist, producing architecture/nfr-spec.yaml in the expected STM location.

Quick Start

Ask the tech-architect agent to run infer-nfr-spec-from-code during /codify using scan-index.json and the prior quality, physical, and logical architecture proposal paths so it can generate architecture/nfr-spec.yaml with fully grounded evidence and resolution traces.

Frequently Asked Questions about infer-nfr-spec-from-code

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I recover enforced non-functional requirements from existing production code?

The mechanism works by pairing every inferred NFR entry with its delivery mechanism, such as a library or middleware, and a verification method, like a test or CI signal, ensuring contracts are fully grounded in scan-index evidence.

Can I derive performance and reliability contracts without guessing thresholds in a brownfield project?

Yes, you can derive performance, reliability, and security contracts in brownfield projects by applying this inference during the codify pipeline, using no-invention guardrails to record missing signals as knowledge gaps rather than fabricating targets.

How do I generate an nfr-spec.yaml with fully grounded evidence during architecture codification?

Required inputs include validated JSON or YAML files, an LTM resolution protocol run, and evidence-grounded NFR entries mapping each mechanism to its verification method before writing to the provided decision manifest paths.

What happens when code scanning lacks signals for certain quality attributes like observability?

When code scanning lacks signals for quality attributes like observability, the no-invention guardrails record those missing or unsupported signals as knowledge gaps rather than fabricating thresholds or verification steps.

Does NFR inference work without prior physical and logical architecture proposals?

NFR inference requires existing quality, physical, and logical architecture proposals to exist before running, as it integrates during the codify pipeline after these proposals are generated to ensure properly grounded evidence.