paulbaranowski

paulbaranowski/wild-horses

Plugin Claude Code

Plugin marketplace listing 14 plugins: harness, autonomous, linting-hooks, marketplace, pyright.

paulbaranowski/wild-horses

Hook Claude Code

Runs after a tool call finishes for Edit and Write tool calls, executing validate-plugin.sh. From paulbaranowski/wild-horses.

paulbaranowski/wild-horses

Settings file Claude Code

Agent settings declaring 1 hook event (PostToolUse).

paulbaranowski/wild-horses

Instructions file

Instructions for paulbaranowski/wild-horses, covering wild horses - claude code plugin marketplace, project structure, key references, rules and marketplace structure.

paulbaranowski/wild-horses

Agent

How the engineering skills should consume this repo's domain documentation when exploring the codebase.

paulbaranowski/wild-horses

Agent

Issues for this repo live in Linear. Default workspace: HRD (Herds). Default project: Wild-Horses.

paulbaranowski/wild-horses

Agent

The skills speak in terms of five canonical triage roles. This file maps those roles to the actual label strings used in this repo's Linear workspace.

paulbaranowski/wild-horses

Plugin Claude Code

An autonomous skill that drives a single issue/ticket link or plan file all the way to an opened pull request with no human in the loop. Hand it a Linear/GitHub/other issue URL, a path to a plan/spec file, or a plan already in the conversation, and it decides everything itself: implements, tests, simplifies the diff…

paulbaranowski/wild-horses

Skill Claude CodeCodex

Autonomously take an issue/ticket or a plan file from a link (or path) to an opened pull request, with no human in the loop. Hand it a Linear/GitHub/other issue URL, a path to a plan/spec file (e.g. /plans/repo/foo.md), or a plan already read into the conversation, and it decides everything itself: implements, tests…

paulbaranowski/wild-horses

Plugin Claude Code

Reclaim disk space from stale git worktrees. Scans configured roots (direct repos and parent dirs), classifies each worktree as cleanable (merged PR, upstream gone, merged into the default branch, or stale) or skipped (dirty, locked, unpushed), shows the cleanable set grouped by reason with sizes, and removes the ones.

paulbaranowski/wild-horses

Hook

Runs before the agent uses a tool for Bash tool calls, executing cleanup-worktrees-cli-allow.sh via bash. From paulbaranowski/wild-horses.

paulbaranowski/wild-horses

Skill Claude CodeCodex

Reclaim disk space from stale git worktrees: scans configured roots, classifies each as cleanable (merged PR, upstream gone, merged into the default branch, stale) or skipped (dirty, locked, unpushed), shows the cleanable set grouped by reason with sizes, and removes the ones you pick. Use when the user asks to clean…

paulbaranowski/wild-horses

Plugin Claude Code

Map and visualize codepaths in any codebase as an interactive architecture diagram.

paulbaranowski/wild-horses

Skill Claude CodeCodex

Scan a codebase and produce architecture.json + codepaths.json describing components, edges, and traced codepaths. Renders architecture.html at the end. Use when the user says "map my app", "build the architecture viz", "scan my codepaths", "map the X flow", etc. Pass --render-only to skip all mapping work and just…

paulbaranowski/wild-horses

Skill Claude CodeCodex

Open docs/codepaths/architecture.html in the browser. With --select, blocks on a user pick and returns the chosen codepath JSON to the calling agent. Run after /codepath-mapper has produced the HTML.

paulbaranowski/wild-horses

Plugin Claude Code

Analyze code for feedback-loop blockers and AI reasoning gaps, then build, run, and inspect a structured remediation task list to fix what it finds. Ships two analysis commands (/harness:feedback-blockers, /harness:reasoning-gaps) and three task-list skills (task-list-builder, task-list-runner, task-list-viewe.

paulbaranowski/wild-horses

Command

Analyze code for feedback-loop blockers — encapsulation gaps, OOP design issues, testability barriers, and harness-unfriendly patterns that prevent fast, clear change-test-fix cycles. Spawns 4 parallel specialist agents, merges findings, and produces a prioritized remediation plan. Use when code changes cause…

paulbaranowski/wild-horses

Command

Analyze code for AI reasoning gaps — untyped signatures, implicit control flow, hidden state, missing docs, and structural complexity that prevent agents from tracing data flow and predicting behavior. Spawns 3 parallel specialist agents, merges findings, and produces a prioritized remediation plan. Use when AI agents…

paulbaranowski/wild-horses

Skill Claude CodeCodex

Build or rewrite a structured task list (JSON + paired markdown report) matching the harness task-list schema. Accepts free-form text, an existing reasoning-gaps/feedback-blockers report, an existing JSON task file (in-place rewrite), or recent conversation context. Use when the user says "build the task list using…

paulbaranowski/wild-horses

Skill Claude CodeCodex

Run a structured task list (JSON file in the harness task-list schema) by dispatching each task to a foreground Agent in sequence. Auto-locates an in-progress task file in docs/exec-plans/active/ when no path is given. Use when the user says "run the plan", "resume the plan", "execute the tasks", "run…

paulbaranowski/wild-horses

Skill Claude CodeCodex

Read-only viewer for harness task-list JSON files. Auto-locates the active task file in docs/exec-plans/active/. With no args, shows counts, the paired markdown plan path, the in-progress task ID (if any), and pending task titles. With a positional task ID, prints that task's full details. Use when the user says "show…

paulbaranowski/wild-horses

Plugin Claude Code

PostToolUse hooks that run Prettier + markdownlint on .md edits and pyright on .py edits. Hooks are registered automatically; /linting-hooks:install detects and installs the required software per machine.

paulbaranowski/wild-horses

Command

Install the software the linting-hooks plugin's PostToolUse hooks depend on (prettier, markdownlint-cli2, pyright, jq). Hooks are registered automatically on plugin enable; this command only manages the per-machine software each hook needs to actually do work.

paulbaranowski/wild-horses

Skill Claude CodeCodex

Scaffold a Claude Code plugin marketplace with proper structure, schema validation, and CLAUDE.md conventions. Asks for the marketplace name and whether to import an existing skill. Use when starting a new marketplace repo or adding marketplace structure to an existing repo.