j-mauro-r/SAT-Dengue-V2 / .agents/skills/deepworkplan
deepworkplan
VerifiedDeepWorkPlan — turn any repo AI-first and run Deep Work Plans. Routes to create, execute, refine, resume, status, verify, upgrade, and repo-onboarding sub-skills based on intent. Use when the developer wants to plan, execute, manage, or verify structured multi-task work, or make a repository AI-agent-ready.
shared/adaptation.md
View on GitHub# adaptation.md — Reasoning Over Copy-Paste
> How DeepWorkPlan sub-skills (especially `onboard`) adapt to a target
> repository instead of blindly copying a template.
## The principle: reason, don't copy-paste
When a flow generates AI-first structure (`AGENTS.md`, `docs/`, per-module docs,
`.agents/`, validation commands, example plans) for a target repo, it **MUST
reason about that repo** rather than copy-pasting a fixed template. The split
comes from an audit of six Dailybot repositories (`../spec/README.md`): **most
of the AI-first structure was common across them, and a small remainder was
repo-specific** — the validation commands, paths, primary stack,
stack-specific skills and example plans. That remainder is exactly the part
that must be **reasoned per repo**. The proportion is an observation about six
repositories, not a measured constant; treat it as direction, never as a
figure to quote.
Concretely, before writing anything, a flow **MUST** discover:
- The repo's **primary stack** (language, framework, package manager) from
manifest files (`package.json`, `pyproject.toml`, `go.mod`, `Cargo.toml`,
etc.).
- The repo's **validation commands** (lint, type-check, test, build) from
scripts/config — never assume a fixed command set.
- The repo's **architecture and module layout**, to place per-module docs where
they belong.
- Any **existing** AI-first setup, so generation **reconciles** with it rather
than clobbering it (approval required before destructive changes).
The common part (the baseline shape — `AGENTS.md` as index + rules + quick
commands, the `docs/` categories, `.agents/` with the `.claude → .agents` and `.cursor → .agents`
symlinks, the `.dwp/` output convention) is applied as-is; the repo-specific
part is filled by reasoning.
## The two archetypes
Adaptation also forks on the repository archetype (see Task 2's
`../spec/ARCHETYPES.md`):
- **Individual repo (the common case)** — one primary stack. The lean path:
classify → reason about the stack-specific part → generate the baseline.
- **Orchestrator hub** — a coordination repo orchestrating multiple
sub-repositories. The additive path: baseline first, then layer hub-only
structure (sub-project navigation index, `ECOSYSTEM_CONTEXT.md`, cross-project
standards, repository-boundary rules) and wire the orchestrator/child-DWP
capability.
The `onboard` flow classifies the target repo using the `../spec/ARCHETYPES.md`
heuristic and confirms with the user when signals are ambiguous.
> Deeper per-stack presets and the full onboarding reasoning procedure are
> elaborated in **Task 5** (`onboard/SKILL.md`). This stub fixes the principle
> and the archetype fork.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |