Convert legacy requirements.md into a delta-based technical roadmap with investigation artifacts.
npx skills add https://github.com/sandeco/reversa --skill reversa-plan-sandeco
Please help me install this Agent Skill. Skill: reversa-plan Source: https://github.com/sandeco/reversa/tree/main/agents/reversa-plan Command: npx skills add https://github.com/sandeco/reversa --skill reversa-plan-sandeco
It converts a feature’s requirements.md into a concrete technical plan expressed as a delta against what already exists in the legacy codebase, so AI agents can implement changes safely and traceably.
roadmap.md that summarizes approach, architectural delta, data delta, contract delta, migration plan, risks, and “criterion of pronto”._reversa_sdd/* artifacts plus .reversa/principles.md) and prepares an impact list that will feed later /reversa-coding.interfaces/<nome>.md files per affected external contract (HTTP, queue, gRPC, GraphQL) including request/response, errors, idempotency, and timeouts.[DÚVIDA] markers in requirements.md, routes clarification via /reversa-clarify, and—if confirmed—turns each doubt into explicit roadmap premises.Run the skill by asking your agent: “/reversa-plan”.
High-intent search queries and answers about installing and using this skill.
To generate a technical roadmap delta, translate a legacy `requirements.md` into a delta-based plan that summarizes architectural, data, and contract changes against the existing codebase. This produces a `roadmap.md` file that AI agents use to implement modifications safely and traceably.
Doubt-aware planning detects `[DÚVIDA]` markers inside `requirements.md` during legacy modernization. It routes these items to a clarification process and, once resolved, converts each confirmed doubt into explicit premises within the final technical roadmap.
Yes, you can generate external interface contracts for HTTP, gRPC, GraphQL, and queues. The process creates individual `interfaces/<nome>.md` files per affected contract, detailing request/response schemas, errors, idempotency rules, and timeouts.
The best way to plan data schema changes is by producing a `data-delta.md` artifact alongside the technical roadmap. This data delta maps structural modifications against the existing legacy state, ensuring AI agents understand migration requirements and risks.
Yes, delta-based technical planning requires reading `.reversa/state.json` and `.reversa/active-requirements.json` to validate active requirements. It also applies `before-plan` hooks and reads `.reversa/principles.md` to establish context.
cliffthelinGenerate clarifying questions and update requirements.md with user answers.
JeffallanAutomate legacy-system modernization with strangler fig and branch-by-abstraction patterns.
ricardops34Generate targeted questions and integrate answers into requirements.md.
stribusTranslate legacy requirements into a technical delta roadmap with artifacts.