ilude/dotfiles - Page 2

logging-observability

Logging, metrics, tracing, monitoring, alerting, SLOs, correlation/trace IDs, or log aggregation. Not for debugging without telemetry.

m365-tenant-automation

M365 scripting: Entra ID, Exchange, SharePoint, Teams, Graph, licensing, inventory, or tenant config. Not for unrelated PowerShell.

no-ai-slop

Prose cleanup for filler, hype, vague claims, repetition, detection tells, or uncited specifics. Not for Markdown architecture or archival work.

orchestration

Coordinate bounded root-to-coordinator-to-leaf work when independent execution, verification, or context isolation provides a concrete benefit.

pdf-reader

Read, search, extract, or render PDFs, including page-specific or visual questions.

pi-command

Pi slash commands, prompt templates, workflow skills, or command placement. Not for skill authoring; use skills-engineer.

pi-contributor-workflow

Only for badlogic/pi-mono or @mariozechner/pi-* upstream issues/PRs, approval, contributor gates, pkg:* labels, changelogs, or accepted fixes.

pi-extension

Pi extensions and tool policy: extensions/*.ts, hooks, registerTool, promptGuidelines, registerCommand, footer/status UI, tool_result, session hooks, or subprocesses. Not for slash-command placement; use pi-command.

pi-goal

Write or improve Pi /goal, goal prompt files, long-running objectives, or goal_prompt_file.md. Not for PRDs or task planning.

pi-log-analytics

Pi session, trace, metrics, workflow-friction, workflow-telemetry, usage, or local JSONL analysis with DuckDB. Use the registered catalog and bounded SQL query interface for Pi runtime logs. Not for adding telemetry, generic SQL/database design, or non-Pi…

planning

Requirements, user stories, acceptance/verification criteria, testable outcomes, or wayfinding for a large unclear effort across sessions. Not for PRD drafting (/prd) or Pi /goal prompts.

playwright-e2e

Playwright specs, fixtures, traces, browsers, test projects, gated suites, or e2e Docker cleanup. Not for unit-test-only changes.

powershell

PowerShell syntax and scripts: .ps1, cmdlets, modules, script blocks, pipelines, quoting, interpolation, native commands, error semantics, or PowerShell 5.1/7 compatibility. Not for Bash/POSIX shell; use shell.

prd

Use to draft, refine, or review a Product Requirements Document. Not for acceptance criteria, planning, or Pi /goal prompts.

private-store

Save, archive, pack, encrypt, or commit sensitive local private/ files with Dolos.

python

Python files, pyproject.toml, Python packaging, pytest, or Python-specific runtime behavior. Use for Python syntax, imports, exceptions, resource cleanup, and compatibility traps. Not for general repository policy.

reddit

Reddit public .json search, subreddit listings, or a specific post.

ruby

Ruby/Rails: .rb, Gemfile, Rakefile, .gemspec, bundler, ActiveRecord, migrations, RSpec, minitest, or generators.

rust

Rust: .rs, Cargo.toml, Cargo.lock, ownership, borrowing, lifetimes, traits, cargo build/test/run, Result/Option, or borrow checker.

shell

Bash/POSIX shell: .sh, .bash, .zsh, Makefile fragments, CLI scripts, quoting, exits, traps, or cross-platform shell behavior. Not for PowerShell; use powershell.

skills-engineer

SKILL.md, skill frontmatter/descriptions, activation triggers, agent definitions, or command/agent instructions. Not for Pi slash-command placement; use pi-command.

ssh

SSH client/server configuration and troubleshooting: ssh, sshd, keys, agents, known_hosts, authorized_keys, AuthorizedKeysCommand, ProxyJump, authentication, or host-key verification. Not for Git operations; use git-workflow.

terraform

Terraform: .tf, variables.tf, main.tf, outputs.tf, terraform.tfvars, .tfstate, HCL, init/plan/apply, or remote backends.

test-review

Reviews JavaScript and TypeScript test-suite value, protection, reliability, and maintainer cost. Use for explicit /test-review baseline, diff, path, deep-performance, or smells requests. Not for general code review, browser-E2E semantics, or remediation…

testing

Cross-language unit and integration testing: fixtures, isolation, teardown, mocks, determinism, coverage, and test artifact cleanup. Use with a language skill for runner-specific syntax. Not for browser E2E; use playwright-e2e.

tui-ux

Interactive terminal UX: command palettes, pickers, setup wizards, lists, forms, keyboard navigation, or streaming state. Not for passive text formatting, tool-call/result rendering, or status labels unless interaction changes.

typed-agent-workflows

Pi defineAgent, typed semantic stages, or code-plus-agent workflows using pi/lib/typed-agent.ts. Not for delegation, extension hooks, or command UX.

typescript

TypeScript or JavaScript files, package manifests, workspaces, module boundaries, or configured test/build behavior. Use for TypeScript-specific runtime and package traps. Not for general repository policy.

ux-design-workflow

Web or graphical UX/UI: accessibility, design tokens, aesthetics, responsive behavior, or design pipelines. Not for terminal interfaces; use tui-ux.

war-report

Generate war reports, weekly activity reports, or work accomplishment summaries from Git commits.

workflow-design

Command UX: command surfaces, task runners, package scripts, Makefile/just workflows, deployment commands, or workflow docs. Not for shell syntax or Justfile implementation.

bad-name

---

Source text: Undetermined

clean-skill

Handles focused clean fixture reviews.

long-body

Exercises long body detection.

missing-boundary

Helps with missing boundary examples.

missing-reference

Exercises missing reference detection.

trigger-overlap-one

Reviews overlapping docker deployment workflows.

trigger-overlap-two

Reviews overlapping docker deployment checks.

analysis-workflow

Analysis, diagnosis, adversarial critique, or coding mechanism selection, including routine edits. Not for diff review, architecture design, planning, or routine edit consistency.

architecture-design

Module, interface, seam, dependency, or structural design. Not for approach selection, planning, or routine edit consistency.