hackify-edge

Plugin hackify-edge from marketplace hackify-marketplace in nadyshalaby/hackify. Commands read from the repo's manifest — 1 month ago.

Skills (7)

• codewalk

  DEEP-BY-DEFAULT browser-based interactive call-stack viewer for code the user did not write, a senior-peer walkthrough of ONE execution path from a single entry point (route, handler, CLI command, queue job, UI action), rendered as a GitHub-PR-style three-pane app under `.codewalk/<slug>/` at the re

• groom

  Socratic pre-task refinement skill, the idea-shaping front door to the hackify workflow. Runs an interactive 1-or-2-forking-questions-per-turn conversation (NOT a batched wizard), reflects what the user said before each following question, and graduates to quick, the default build route, the moment

• hackify

  One unified end-to-end dev workflow for ANY substantive task, feature, bug fix, refactor, redesign, design, debug, migration, or research-then-build. Driven by one markdown work-doc at <project>/docs/work/, with every clarifying question asked up front in one batched questionnaire, a hard gate befor

• lawkeeper

  Audit an existing or new codebase against its engineering laws, hard caps (function/file/param/nesting), zero-tolerance bans (lint suppressions, non-null `!`, empty catches, bare Error throws, secrets, inline types in scoped modules), plus DRY, naming, layering, single-responsibility, folder-structu

• quick

  The default route for any substantive prompt, a compressed flow that keeps every guarantee of the hackify workflow and drops only the ceremony (Plan+Gate, Spec review, 4-options finish) that costs more wall-clock and tokens than it returns. Auto-discovery triggers, invoke this skill on any prompt ca

• review-triage

  Structured response engine for reviewer findings, converts a batch of comments into a per-finding decision table with columns Finding / Severity / Decision / Evidence, where every row picks exactly one of `accept` / `push-back` / `defer` / `needs-restatement`, the last of those being the one value t

• skillsmith

  Authors NEW skills that conform to hackify's binding contracts. NOT a generic Claude Code skill creator. Produces SKILL.md files under skills/<slug>/ that pass every check the validate-dod.sh harness runs against hackify-conformant skills, the 8-section sub-agent contract, the 4-section Wizard contr

Slash commands (2)

• /hackify-edge:designify

  Author, extract, or refresh the project's design spec at docs/design/DESIGN.md plus its visual preview.

• /hackify-edge:summary

  Print the plain-language update log for the current hackify task, one block per change (what was wrong, why it happened, what I did, how I know it works, status).

Subagents (10)

• codebase-investigator

  Read-only investigation agent serving two phases from one prompt, with mode picking the lens. In research mode (Phase 1) it answers ONE load-bearing question about an unfamiliar codebase, builds keyword sets from the question, greps inside the search scope, discards hits the dispatcher already ruled

• finding-refuter

  Phase 5 adversarial refuter, judges reviewer findings BEFORE the address-all loop spends a fix on them. Takes every finding of a review round verbatim and returns one verdict each, UPHELD (claim survives, cite the reachable entry point or the real rule), REFUTED (cite the file:line guard, carve-out,

• implementer

  Phase 3 implementation worker, the one agent type every Phase 3 dispatch takes: a solo foundation wave, a concurrent module track, a solo assembly wave, a testing wave that runs solo or splits into concurrent ones, a single-track round and a quick-mode change. Produces a minimal, test-anchored diff

• reviewer-coherence

  Phase 5 Multi-reviewer F, audits a base..head git diff for cross-module coherence defects, the discrepancies that appear because a wave's implementer is blind to every wave that ran before it and to every line of pre-existing code. Compares every boundary-crossing symbol's producer against every con

• reviewer-design

  Phase 5 Multi-reviewer E, audits a base..head git diff for design-conformance defects against the project's committed docs/design/DESIGN.md (hardcoded color/size/shadow literals where a token exists, off-ramp type sizes, component state gaps, violations of the spec's own Don'ts list, WCAG AA contras

• reviewer-performance

  Phase 5 Multi-reviewer D, audits a base..head git diff for performance defects (N+1 / query-in-loop, algorithmic complexity on hot paths, unbounded growth in caches/results/fan-out/listeners, wasted parallelism, blocking sync I/O on servers, re-render storms and layout thrash, missing pagination/bat

• reviewer-quality-plan

  Phase 5 Multi-reviewer B, the panel's floor, carrying two lenses over one read of the diff. Quality and layering, audits a base..head git diff for DRY violations against existing helpers, function/parameter/nesting/file size caps, inline types in forbidden files, new lint suppressions, non-null asse

• reviewer-security

  Phase 5 Multi-reviewer A, audits a base..head git diff for security & correctness defects across all ten domains of the catalog it cites (auth flows, permission boundaries, injection, PII/secrets, migration safety, race conditions, CORS and SSRF, cryptography, supply chain, error handling), citing h

• reviewer

  Phase 5 all-lens reviewer, the explicit, named, lower-cost opt-out from the five-agent panel that is Phase 5's default reviewer route in both quick and full mode, reached by a user asking for it directly rather than by any demotion the parent applies on its own. One agent carries all five review len

• spec-reviewer

  Phase 2.5 spec reviewer, the single agent that audits a hackify work-doc before Phase 3 implementation begins. Carries three lenses over one read. It audits internal consistency (Q&A vs Approach vs DoD vs Sprint Backlog contradictions, unaddressed Original Ask sentences, DoD bullets without covering

Hooks (8)

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}/hooks/inject-context.sh" --event SessionStart "${CLAUDE_PLUGIN_ROOT}/rules/plugin-map.md"

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}/hooks/inject-context.sh" "${CLAUDE_PLUGIN_ROOT}/rules/hard-caps.md"

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}/hooks/inject-context.sh" "${CLAUDE_PLUGIN_ROOT}/rules/expert-mindset.md"

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}/hooks/inject-context.sh" "${CLAUDE_PLUGIN_ROOT}/rules/perf-guardrails.md"

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}/hooks/inject-context.sh" "${CLAUDE_PLUGIN_ROOT}/rules/phase-discipline.md"

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}/hooks/inject-context.sh" "${CLAUDE_PLUGIN_ROOT}/rules/claim-integrity.md"

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}/hooks/block-banned-tokens.sh"

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}/hooks/block-ai-attribution.sh"

Similar plugins