words

A generated vocabulary reference for Montology, defining terms used by its ontology, scanner, and code checks.

In plain words
What is it for?
Use it to look up defined terms, understand their namespaces and meanings, and identify collisions or missing vocabulary.
Why use it?
It explains how the project’s words are scoped and prevents code names from conflicting with established meanings.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/shinyobjectz/montology/words
Any agent
npx skills add shinyobjectz/montology --skill words
Clone the repo
git clone --depth 1 https://github.com/shinyobjectz/montology

Made for: Claude Code, Codex.

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,808 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.02808
Opus 5 $0.00024 $0.01404
Sonnet 5 $0.00010 $0.00562
Haiku 4.5 $0.00005 $0.00281

Measured yesterday against content hash 3b95ec16ba20, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

words scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/words/SKILL.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GENERATED FILE. The database at .monty/ontology.db is the truth; this renders from it. Hand edits are lost on the next sync — change the database instead (monty onto add), which is also what monty lint verifies.

The words

word code kind is test
amendment (inside ontology) onto.amendment core · noun a word's own text corrected in place — the name and its history stay, and what the field said before is ledgered; the counterpart to a rename, which retires the name instead what a word used to say
candidate (inside scan) scan.candidate core · noun a recurring declared name with no word — vocabulary the codebase is asking for what wants a definition
code (inside ontology) onto.code core · value a word's dotted address (har, har.cell) — prefixes must resolve to words, so the namespace stays a tree where a word lives
collision (inside scan) scan.collision core · noun a declaration named after a word that means something else — the failure scan exists to catch where code and vocabulary disagree
convergence (inside scan) scan.convergence core · noun the state a tended lexicon reaches: new concepts per sample decaying toward zero while the code keeps growing whether the vocabulary settles
disclosure (inside ontology) onto.disclosure core · noun how much of the vocabulary a render keeps resident and how much it points at — the always-loaded page, the reference pages behind it, the database behind those what every agent is made to carry
divergence (inside scan) scan.divergence core · noun one word declared as more than one value — the finding no exception silences, because sharing a name is a decision and meaning two things is not where one noun holds two things
doctrine (inside ontology) onto.doctrine core · noun a decision written down in the database — because a decision that is not written down gets re-litigated why it is this way
drift (inside scan) scan.drift core · noun a value or prose that moved away from the vocabulary it claims: a rogue literal, a near-duplicate, a stale render what the gate exists to catch
exception (inside ontology) onto.exception core · noun a recorded decision that a symbol may share a word's name — with the reason it was granted and the paths it holds in, because a reasonless allow-list is one nobody reads which collisions we keep
gist (inside ontology) onto.gist core · noun a definition rendered to its first sentence — what a resident page carries when the whole definition is not worth its place in context the short form of a meaning
glossary (inside ontology) onto.glossary core · noun the whole ontology rendered to one page — every word, ruling and doctrine block, from the database, with the intake it grew from as appendix what our words mean, on one page
intake (inside workspace) ws.intake core · noun the phased questions a workspace starts with — asked in a form, answered on disk, each round written from the last — before any word is authored what was asked, what they said
migration (inside scan) scan.migration core · noun a renamed word propagated through the code by token — sweep reports, apply rewrites, the diff is the review how the code catches up
ontology onto core · noun a repo's vocabulary as a database: words, doctrine, rulings — enforced by scan, rendered to prose, never authored in prose what the words are
phantom (inside surface) surf.phantom core · noun a surface with no seam: declared, never met — the mirror of a candidate, which is a declaration with no word what nothing touches
recipe (inside ontology) onto.recipe core · noun a named composition of utility classes — the unit a utility-first design vocabulary trades in above single tokens what a repeated class string wants to be
ruling (inside ontology) onto.ruling core · noun a recorded boundary decision: an overload (say Y not X), a collision (whose word it is, who moved), or a rename (was, now, why) how arguments end
scan scan core · noun the tree-sitter sweep of a codebase: every declaration measured, every dependency surface and the seams between them, checked against the vocabulary what the code claims
seam (inside surface) surf.seam core · noun one point where two surfaces meet — an import that resolves, a call that lands, a config key read; direction is an attribute, so inputs and outputs need no words of their own where two things touch
surface surf core · noun what a thing exposes: its named, callable, importable face — ours and a dependency's alike, whose it is being an attribute and not a second word what something offers
sync (inside ontology) onto.sync core · verb rendering the database to the generated words skill — prose is output, never source how agents read it
token (inside ontology) onto.token core · noun a named design value — color, space, radius, shadow, font, breakpoint — the vocabulary a visual system trades in what design means
vitals (inside scan) scan.vitals core · noun the at-a-glance state of a repo's meaning: the gate, the vocabulary, the design system, the guard — one verdict (tended, drifting, untended) how the repo is doing
word (inside ontology) onto.word core · noun one term with one meaning: definition, the one-line test, an optional owner and dotted code what we mean
workspace ws core · noun any repo montology is initialized into — found by walking up for .monty, the way git finds .git where work happens

Read the full file on GitHub · 89 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 89 lines · 48 tokens per session scan A 3b95ec16ba20

Subscribe to this mod's changes

words is a skill published in the GitHub repository shinyobjectz/montology (1 stars, last pushed 7d ago), licensed Apache-2.0. It adds 48 tokens to every session and 2,808 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.