syv-skills

Plugin syv-skills from marketplace syv-skills in syv-ai/agentic-coding-playbook. Commands read from the repo's manifest — 2 days ago.

Skills (22)

• add-visual

  How to add or restyle a D3 diagram in the playbook docs (MkDocs Material). Use for any flowchart, diagram, chart, or graphic in docs/.

• knowledge

  Add or update a playbook knowledge-base source. Review the source, choose a destination, scan for contradictions, and draft caveated entries.

• brainstorming

  Turns an idea into an agreed design through collaborative dialogue — purpose, constraints, approaches and trade-offs — before any code is written. Use when a feature, component or behaviour change is underspecified, when several approaches are plausible, or when the user wants to think a design thro

• executing-plans

  Executes a written implementation plan task by task, verifying each step and reviewing the result against the plan before calling it done. Use when there is an existing plan to implement, in this session or one started fresh from a handoff.

• grill-me

  Interviews the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when the user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

• guardrails

  Reviews what a coding agent could damage in this repo, notes the controls already in place, and asks which remaining risks to cover.

• handoff

  Compacts the current conversation into a handoff document so another agent can pick up the work. Use when you want to hand off to a fresh session, or when continuing in a clean context would work better than carrying this one forward.

• improve-code-design

  Find named design problems in a codebase — the anti-patterns coding agents produce most — and name the remedy for each. Use for a design review, when asked whether code violates SOLID/DRY/YAGNI, when assessing agent-generated code, when someone says changing one thing keeps breaking another, or when

• improve-codebase-architecture

  Find deepening opportunities in a codebase — refactors that turn shallow modules into deep ones — and present them as a visual HTML report. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-na

• prototype

  Builds a throwaway prototype that settles a design question before committing to it — a self-contained HTML demo for state and business logic, or several radically different UI variations on one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, ex

• receiving-code-review

  Evaluates code review feedback against the codebase before acting on it, with clarification, evidence-based pushback and a check against over-building. Use when receiving review comments from a person, a reviewer agent or a review tool, especially when feedback is unclear, broad, or seems technicall

• retro

  Reviews a session that was harder than it should have been and proposes changes to the agent's environment so the next run goes better.

• review-tests

  Critically reviews the design of a test suite — tests that pin a poor production design, near-duplicate tests that reveal duplicated code, tests that freeze an over-specialised design, and tests that keep dead code alive — and reports each with a named remedy and a handoff file. Use after an agent h

• setup-syv-skills

  Interviews the team about how agents should work in this repo, records the answers in the project's instructions file, and offers to add a commit gate.

• systematic-debugging

  Root-cause debugging for bugs, test failures, flaky tests, build failures, performance regressions and unexpected behaviour, built around a fast reproducible feedback loop and runtime evidence. Use before proposing a fix whose cause isn't already clear, when something passes locally but fails in CI,

• tdd

  Test-driven development with a red-green-refactor loop and tracer-bullet vertical slices, testing behaviour through public interfaces rather than implementation. Use when implementing a feature or bugfix that has a testable seam, when the user mentions TDD, red-green-refactor or test-first, or wants

• track-work

  Records what is being built and breaks it into independently grabbable work items in the project's tracker.

• visual-plan

  Renders an implementation plan as a single self-contained HTML review page — diagrams, file map, annotated code and open questions — published as a private page or opened as a local file, with decisions captured back in the conversation. Use when a plan is multi-file, architecture-, data- or UI-heav

• work-autonomously

  Carries a plan, spec or PRD through to verified, working code without check-ins, logging each decision in design-decisions.md.

• write-a-skill

  Writes and revises agent skills (SKILL.md folders) — whether one is worth writing, a description that triggers, a body agents follow, bundled scripts and references, and how to test it. Use when the user wants to create, edit, review or test a skill, when a skill doesn't trigger or isn't followed, o

• writing-plans

  Writes an implementation plan from a spec or agreed requirements — files to touch, ordered tasks, and how to verify each step. Use when a change spans several files or steps, when the approach needs to be pinned down before coding, or when another session or agent will do the implementation.

• zoom-out

  Maps an unfamiliar area of code one level of abstraction up — the relevant modules, their callers, and how data flows between them. Use when you're unfamiliar with an area of code, need to see how it fits into the bigger picture, or are about to change code whose surroundings you don't know.

Similar plugins