hskills-core

Plugin hskills-core from marketplace hskills in Hissal/hskills. Commands read from the repo's manifest — yesterday.

Skills (44)

• architecture-review

  Scan a codebase for deepening opportunities, present them as a visual HTML report, then stop for your pick.

• blast-radius

  Find what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it's safe because of by running real code instead of writing it up. Use for 'blast radius of X', 'what could this break', or reviewing a small diff you don't trust.

• break-ui

  Exercise a UI with realistic worst-case data and failure conditions, and report what breaks before anyone fixes it.

• code-review

  Use when the user wants code changes reviewed for problems: a branch, a pull request, or their work in progress, such as before opening a PR.

• codebase-design

  Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

• commenting

  Use when posting a comment, reply or dismissal reason on a PR or issue on the user's behalf.

• constraints

  Turn the project's quality bar into enforceable checks, through a short interview with a default for each question; place each check by cost, rank it by how hard it is to fake, ratchet measured values, and put a floor guard into the project that fails a change reaching green the cheap way.

• context-audit

  Audit everything the agent is told across its context layers, measure what it costs, and propose a diff without applying it.

• delegating

  Use before dispatching any subagent or delegated agent, or when choosing the model, harness or effort one runs on.

• dependency-audit

  Audit dependency and supply-chain risk in the project's own package manager, and report what it ships.

• diagnosing-bugs

  Use when the user says "diagnose" or "debug this", or reports something broken, throwing, failing or slow.

• domain-modeling

  Use when discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR.

• grill-with-docs

  A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

• grilling

  Use when the user asks to grill, stress-test or poke holes in a plan, decision or idea.

• guide

  Find the next workflow for your task.

• handoff

  Compact the current conversation into a handoff document for another agent to pick up.

• hillclimb

  Improve one measured result, such as load, build or test time or allocations, through one-at-a-time keep-or-revert experiments with a committed log.

• how

  Use for \"how does X work\" in the local codebase or an installed package, code walkthroughs before changing something, and placement / ownership / layering questions (\"where should this live\", \"which package owns this\", \"is this the right layer\"). Explains subsystem architecture, runtime flow

• implement-spec

  Implement the result of /to-spec and /to-tickets in code.

• implement

  Implement a piece of work based on a spec or set of tickets, then open a pull request for it.

• migrating

  Rename, move, replace or retire something others depend on without breaking its consumers.

• pr

  Use when writing or updating a PR body, pushing to a PR, or merging one.

• prototype

  Build a throwaway prototype to answer a design question. Use when the user wants to check a state model or logic, explore what a UI should look like, or try how something works in use.

• receiving-review

  Use when acting on review feedback (PR review comments, a review workflow's findings, a bot's flagged issue) or when asked whether a finding is real.

• refactoring

  Use when restructuring code while its behaviour stays the same: a rename, move, extract or split, including tidying mid-task when nobody asked.

• research

  Use when the user wants a question researched from external primary sources, such as official docs, specs or upstream source code, and the findings written up.

• retro

  Conduct a retrospective on recent coding sessions and their reviews.

• reviewing

  Use when a review's scope needs settling before reviewing: exactly which changes a branch adds over its merge target, only this session's changes, or which modules changed most recently and deserve review.

• risky-changes

  Use when a change changes a public API, transforms or drops data, or changes pricing, quotas, a default, or a provider or model parameter.

• second-opinion

  Get a fresh second read of a diff, claim or decision, preferably from another model family.

• setup-project

  Set up or re-check a project's configuration for the skills, in an empty, partly set up or established repo.

• tdd

  Use when writing, changing or fixing any test, including building features or fixing bugs test-first (red-green-refactor). Not for reviewing a test suite's quality.

• test-review

  Report on the quality of existing tests, finding weak, cheating and missing tests.

• to-questionnaire

  Turn a decision you can't fully answer into a questionnaire for someone else to fill in.

• to-spec

  Turn the current conversation into a spec and publish it to the project issue tracker: no interview, just synthesis of what you've already discussed.

• to-tickets

  Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker (edges as text in one file per ticket locally, or native blocking links on a real tracker).

• triage

  Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.

• unslop

  Make human-facing prose read plainly and drop AI writing tells, by drafting, rewriting or only diagnosing it.

• verifying

  Use when a claim that work is done, fixed, passing or faster needs evidence from a run: before you make one, or to confirm an agent's report or a measured number.

• wait-what

  Re-explain the last response more clearly and concisely, adding context where it is missing.

• wayfinder

  Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.

• why

  Explain why code ended up this way, from git history and the issue tracker.

• wizard

  Write an interactive bash script that walks you through steps only you can perform.

• writing-for-agents

  Use when creating or editing a skill, AGENTS.md or CLAUDE.md, or any doc an agent reaches by a pointer, such as coding standards, a code map or an issue body an agent works from.

Similar plugins