Davide Magli
Community@edivad1999 · Bologna
Davide Magli maintains a 50-skill agent playbook registry covering Kotlin/Android engineering, adversarial code review, verification harnesses, and software design principles.
Agent Skills by Davide Magli
Showing 45 vetted skills indexed across 1 GitHub repositories.
edivad1999figure-it-out
Designs an auditable multi-phase playbook for large migrations and cross-cutting changes.
edivad1999setup-stuc
Configures per-role model assignments and verifies upstream Android skill stack installations.
edivad1999bro
Restates the previous AI message in plain language without jargon.
edivad1999interrogate
Spawns multiple LLM reviewers to adversarially review code changes and synthesizes a verdict.
edivad1999principle-boundary-discipline
Concentrates validation and error handling at system boundaries while keeping business logic in pure functions.
edivad1999swarm
Orchestrates parallel cloud workers and aggregates their results into one consolidated report.
edivad1999principle-model-the-domain
Encodes domain rules in data structures instead of scattered conditionals.
edivad1999technical-writing
Applies Diátaxis, Google style, STE, and Global English rules to technical documentation.
edivad1999principle-fix-root-causes
Guides debugging by tracing symptoms to root causes instead of applying symptom-level workarounds.
edivad1999why
Investigate design rationale behind code by querying git history and MCP evidence sources in parallel.
edivad1999principle-encode-lessons-in-structure
Converts recurring corrections into lint rules, metadata flags, runtime checks, or scripts.
edivad1999how
Generates architectural explanations of codebases using parallel explorer and explainer subagents.
edivad1999blast-radius
Analyze what a code change could break beyond the diff and prove safety by running real code.
edivad1999principle-guard-the-context-window
Routes large payloads to subagents and keeps summaries in the main context window.
edivad1999principle-subtract-before-you-add
Guides removal of dead code and redundant logic before building additions or refactors.
edivad1999reflect
Spawns three parallel review subagents over a session transcript to route learnings into skill edits.
edivad1999stuc-mode
Applies opinionated Android/Kotlin agent workflows with playbooks, principles, and verification loops.
edivad1999principle-foundational-thinking
Guides structural decisions on data structures, scaffolding order, and concurrency before writing logic.
edivad1999show-me-your-work
Maintains an append-only TSV decision log recording what, why, evidence, and result for reviewable work.
edivad1999principle-build-the-lever
Guides building codemods, scripts, and generators instead of performing non-trivial work by hand.
edivad1999principle-make-operations-idempotent
Designs state-mutating operations to converge to the same end state across crashes, restarts, and retries.
edivad1999using-android-skills
Routes Google Android product and migration workflows to android/skills leaf names.
edivad1999recall
Reconstructs recent working context from chat transcripts and shared records into a status brief.
edivad1999principle-sequence-verifiable-units
Sequences multi-step code changes into small units verified before advancing.
Frequently Asked Questions About Davide Magli
FAQPage SchemaWhat tasks can I accomplish with edivad1999's skill registry?▼
You can design migrations with an auditable playbook, fan out parallel review workers, run adversarial multi-model code review, verify Android UI on emulators, reconstruct working context from chat history, and apply 20+ design principles covering type discipline, idempotency, root-cause debugging, and test-behavior assertions.
Who is the target audience for these skills?▼
Kotlin and Android engineers working with Compose, Flow/StateFlow, coroutines, and KMP boundaries, plus any developer wanting disciplined review, verification, and refactoring playbooks. The stuc-stack skills specifically target users of chrisbanes/skills and the android CLI harness.
How do I install and run the stuc-stack skills?▼
Run /setup-stuc to configure stuc-stack models per role and reasoning budget, doctor upstream installs, and link project-local verify skills. The stuc-chrisbanes glue requires `npx skills add chrisbanes/skills` and fails closed if upstream skills are missing.
What prerequisites do the Android verification skills need?▼
android-verify and verify-notes require Gradle for assemble plus the android CLI binary (describe, run, layout, screen, journeys) and the official android-cli skill; they stop if either is missing. using-android-skills requires the upstream android/skills package for AGP 9, Navigation 3, and CameraX routing.
How do the review and audit skills keep work trustworthy?▼
show-me-your-work logs one TSV row per decision with evidence and results; blast-radius runs real code to prove a change is safe; maintain-verification-skill drives every feature in one live session and ships at most one PR of proven corrections.