.pi/skills/taxonomy-guide/SKILL.md
Naming taxonomy for skills and extensions in this package. Domains, type suffixes, location rules and the decision framework for when to create a new skill, extension or add to AGENTS.md. Use when creating, renaming or reorganizing skills and extensions.
npx skillsauth add jitsusama/agentic-harness.pi taxonomy-guideInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill defines how skills and extensions are named, categorized and organized in this package. Follow it when creating new ones, renaming existing ones or deciding where a piece of guidance belongs.
Every skill name follows the pattern
{domain}-{concern}-{suffix}, where the domain groups
related skills together alphabetically, the concern
describes the specific topic, and the suffix tells you what
kind of skill it is. Some skills have a single-word concern
that makes the domain and concern indistinguishable (e.g.,
planning-guide); that's fine when the domain is the
concern.
Domains are the subject matter. They should all be at the same level of abstraction: nouns that describe what the skill is about, not verbs or activities.
When a skill is about a format standard that isn't tied to a
specific platform, it gets no domain prefix. The absence of a
prefix signals technology neutrality. Platform-specific
formats get their platform domain prefix (github-pr-format,
github-issue-format).
Four suffixes describe what a skill is. Every skill gets exactly one.
When deciding which suffix to use, ask: does this skill tell you how to do something (guide), how to use a tool (convention), how to structure an artifact (format), or what quality looks like (standard)?
Every extension name follows the pattern
{name}-{contract}, where the name describes what the
extension does and the contract suffix identifies its
behavioural contract.
CommandGuardian<T>
interface with detect, parse and review steps. The user
sees the command and can approve, reject or modify it.Every extension must have exactly one suffix. If an extension seems to fit multiple categories, the primary behavioural contract wins. A widget that also intercepts keyboard input is still a widget if its primary purpose is visual.
./skills/: package-bound skills shipped to anyone who
installs this package. These teach general methodology,
conventions, formats and standards that apply across
projects..pi/skills/: project-local skills that only matter
when developing this package. Extension development
guidance, keybinding conventions and this taxonomy itself
live here.The test: would someone installing this package as a consumer
benefit from this skill? If yes, it goes in ./skills/. If
it's about building or maintaining the package itself, it
goes in .pi/skills/.
AGENTS.md describes the project's structure, conventions and design principles. It's always loaded and always in context. Skills are loaded on demand when a task matches their description.
Put it in AGENTS.md when:
Create a skill when:
Skills teach methodology. Extensions enforce it. Some are paired: the planning guide teaches the process; the planning workflow extension enforces constraints and manages state.
Create a skill when:
Create an extension when:
When both are needed, the skill and extension get
complementary names: the skill suffix describes the
guidance type (-guide, -format, -standard), and the
extension suffix describes the behavioural contract
(-workflow, -guardian, -widget).
If the skill is about this package's internals (how to write
extensions, keybinding conventions, taxonomy rules), it goes
in .pi/skills/. If it's about a general practice that any
consumer of this package would benefit from, it goes in
./skills/.
development
Structure of a quest README and the documents that live under it: frontmatter shape, the four core and four optional body sections, emoji glyphs, ID format, alias notation, Cast bullets and Journey entries. Use when writing or editing a quest README, a plan, research, brief or report document under a quest. Pairs with quest-convention for choices like kind, promotion and reordering. Follow the prose-standard for voice.
tools
Operational conventions for the quest system: when to use a quest versus a subquest versus a sidequest, when to scaffold a plan or research document, how to reorder priorities, when to add optional sections, when to conclude versus retire, the resuscitate pattern. Use when driving the quest tool, deciding kind, promoting or parking work, or organising a project as quests. Pairs with quest-format for the on-disk shape.
development
Markdown structure rules: Title Case headings with their exceptions, the line-width target and its legitimate exceptions, reference-style links, fenced code blocks with language tags, tables and lists. Use when writing or editing any markdown file (README, AGENTS, docs, plans, skill files), or when adding a heading, link, table or code block. Owns markdown structure; pairs with prose-standard, which owns voice, grammar, spelling and punctuation.
tools
How to measure whether convention corrections keep recurring in the pi session logs, by category and by week. Use to record a baseline before the convention gates take effect and to re-run afterwards to confirm the recurring categories bend down. Pairs with the convention gates (pr-guardian, issue-guardian, commit-guardian, slack-integration) and the convention-context extension.