MCP Server Builder
anthropics/skills
Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.
NLPM noun/verb registry (R51): pick the canonical term, detect vocabulary drift.
Install Vocabulary
$ npx skills add xiaolai/nlpm --skill vocabulary -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install xiaolai/nlpm vocabulary --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/xiaolai/nlpm.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/nlpm/vocabulary .claude/skills/vocabulary && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "vocabulary" agent skill from https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabulary into .claude/skills/vocabulary/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "vocabulary", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabularyType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add xiaolai/nlpm --skill vocabulary -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install xiaolai/nlpm vocabulary --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/xiaolai/nlpm.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/nlpm/vocabulary .agents/skills/vocabulary && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "vocabulary" agent skill from https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabulary into .agents/skills/vocabulary/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "vocabulary", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add xiaolai/nlpm --skill vocabulary -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install xiaolai/nlpm vocabulary --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/xiaolai/nlpm.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/nlpm/vocabulary .cursor/skills/vocabulary && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "vocabulary" agent skill from https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabulary into .cursor/skills/vocabulary/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "vocabulary", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/xiaolai/nlpm.git --path skills/nlpm/vocabulary--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add xiaolai/nlpm --skill vocabulary -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install xiaolai/nlpm vocabulary --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/xiaolai/nlpm.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/nlpm/vocabulary .gemini/skills/vocabulary && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "vocabulary" agent skill from https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabulary into .gemini/skills/vocabulary/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "vocabulary", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install xiaolai/nlpm vocabularyInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add xiaolai/nlpm --skill vocabulary -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/xiaolai/nlpm.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/nlpm/vocabulary .github/skills/vocabulary && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "vocabulary" agent skill from https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabulary into .github/skills/vocabulary/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "vocabulary", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add xiaolai/nlpm --skill vocabulary -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install xiaolai/nlpm vocabulary --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/xiaolai/nlpm.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/nlpm/vocabulary .opencode/skills/vocabulary && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "vocabulary" agent skill from https://github.com/xiaolai/nlpm/tree/main/skills/nlpm/vocabulary into .opencode/skills/vocabulary/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "vocabulary", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
vocabulary- GitHub stars
- 150
- Token cost
- ~3.9k tokens
- SKILL.md length
- 1,732 words
- Files
- 2
- Skills in repo
- 15
- Repo updated
- First seen
- Licence
- ISC
At a glance
NLPM noun/verb registry (R51): pick the canonical term, detect vocabulary drift.
- Works in 5 steps: Add a term only if it has warrant. The… → Re-run the extraction script. python3… → Slot the term. Add a row to the right… → …
- Agent Workflows work in your project
- SKILL.md covers How this registry is built, Two scopes (P1), Verbs and Nouns, plus 5 more sections
- Calls python3
What it does
Vocabulary is an agent skill from xiaolai/nlpm. NLPM noun/verb registry (R51): pick the canonical term, detect vocabulary drift.
Its SKILL.md is about 3.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `registry.yaml`).
It sits in Agent Workflows. The repository describes itself as: Natural-Language Programming Manager — scan, lint, and score NL artifacts with Claude-native quality scoring. The licence is ISC.
When your agent uses it
- Agent Workflows work in your project
Example prompts
- “/vocabulary”
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
- Add a term only if it has warrant. The four warrant types come from analysis/vocabulary-design-principles.md P6: literary, user…
- Re-run the extraction script. python3 analysis/scripts/extract-vocabulary.py writes analysis/vocabulary-extract/summary.md. New terms…
- Slot the term. Add a row to the right table above (internal verb / auditor verb / artifact-class noun / role-noun / output-class noun /…
- Cite the file evidence. Every row has an "examples" or "file evidence" column. Cite at least one path.
- If retiring a synonym, list it under its canonical verb's "deprecated synonyms" line. Do not silently drop terms; deprecation is itself a…
What it can do on your machine
Read from SKILL.md and the folder at commit bfa2fc2. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
python3
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Vocabulary loads about 3.9k tokens when it runs. Until then it costs about 23 tokens; SKILL.md has 1,732 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from xiaolai/nlpm at commit bfa2fc2, republished under its ISC licence (© xiaolai). 1,732 words, ~3,932 tokens.
Download SKILL.mdSave it as
.claude/skills/vocabulary/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- vocabulary
- description
- NLPM noun/verb registry (R51): pick the canonical term, detect vocabulary drift.
- version
- 0.1.0
- user-invocable
- false
NLPM Domain Vocabulary
The canonical noun-and-verb set NLPM uses to name what it does. Every artifact name, command name, agent name, rule wording, and prose description should draw from this registry. Synonyms are flagged by
/nlpm:checkand penalized by/nlpm:score.Sister skill:
nlpm:conventionsholds upstream Claude Code framework terms (hook events, frontmatter fields, manifest keys). This skill holds NLPM-internal domain language. If a term is from Anthropic's docs, it belongs inconventions. If it is from NLPM's own corpus, it belongs here.
How this registry is built
Every term listed below has literary warrant — it appears in at least one NLPM artifact today, captured by analysis/scripts/extract-vocabulary.py. Re-run that script after adding or renaming artifacts; the freshest output lives at analysis/vocabulary-extract/summary.md. New terms enter the registry only after one of the four warrant types from analysis/vocabulary-design-principles.md (P6) is satisfied — literary warrant alone is the entry bar for terms already in use; new coinages need user, structural, or domain warrant.
Two scopes (P1)
NLPM has two declared scopes. A homonym across them is a boundary, not a collision.
| Scope | Lives in | What it operates on |
|---|---|---|
| internal | commands/, agents/, skills/, bin/, scripts/, project CLAUDE.md | NLPM's own behavior — scoring, checking, fixing, testing, listing, trending NL artifacts in the current project. |
| auditor | .github/workflows/auditor-*.yml, auditor/ | The self-evolution pipeline that audits external repos, opens contribution PRs, tracks outcomes, refines rules. |
A verb that appears in both scopes (scan, test, discover) carries the same identity criterion in both — these are sanctioned homonyms. A verb that appears in only one scope (audit, contribute, track in auditor; score, check, fix, ls, trend, init in internal) is scope-bound; using it in the other scope requires either renaming or declaring a second scope-specific definition.
Verbs
Internal scope (NLPM commands)
| Canonical verb | Output | Judgment? | Examples in corpus |
|---|---|---|---|
score | number + penalty list | no (deterministic) | commands/score.md, agents/scorer.md |
check | consistency violation list | no (deterministic) | commands/check.md, agents/checker.md |
test | pass/fail against named specs | no (deterministic) | commands/test.md, agents/tester.md |
scan | pattern-match findings against a signature database | no (deterministic) | commands/security-scan.md, agents/vague-scanner.md, agents/security-scanner.md |
fix | mutated artifact | no (mechanical) | commands/fix.md |
ls | inventory of artifacts | no | commands/ls.md, agents/scanner.md |
trend | history report | no | commands/trend.md |
init | config file | no | commands/init.md |
Deprecated synonyms (do not use in internal scope):
lint,validate→ usecheck(structural) orscore(quality)find,search,list→ uselsanalyze→ usescore(if quantitative) or pick a more specific verbaudit→ not in internal scope; live in auditor scope only
Auditor scope (.github/workflows/auditor-*.yml, auditor/scripts/)
| Canonical verb | Output | Judgment? | Examples in corpus |
|---|---|---|---|
discover | candidate-repo list | no | auditor-discover.yml |
audit | composite quality + security report | yes | auditor-audit.yml |
contribute | PR(s) to target repo | no | auditor-contribute.yml |
track | outcome events appended to events.jsonl | no | auditor-track.yml |
classify | categorical label on a PR comment | no (model-deterministic) | auditor-classify.yml |
refine | rule-edit PR (human-gated) | no (LLM-generated, human-merged) | auditor-refine-rules.yml |
cite | citation-edit PR (human-gated) | no | auditor-cite-exemplars.yml, propose-rule-citations.py |
diff | finding-by-finding comparison | no | diff-findings.py, auditor-docs-diff.yml |
report | rolled-up summary at a cadence | no | auditor-daily-report.yml, generate-daily-report.py |
review | rule-review issue body (human reads, decides) | yes (human) | auditor-rule-review.yml, generate-rule-review-body.py |
propose | a draft for human approval | no | propose-rule-citations.py |
validate | yes/no on a structural rule | no | validate-feedback.sh, validate-rule-ids.py |
Sibling-granularity note (P1):
reportis broader than its peers (classify,refine,cite,diff). The auditor's noun-named workflow convention absorbs this — workflows that produce reports are noun-named after the output (auditor-daily-report,auditor-rule-review), and the verbreportis the generic act behind that family. Documented inconsistency, not a P1 violation; flagged here so future additions don't replicate the pattern without intent.
Cross-scope homonyms (same meaning in both scopes):
scan— pattern-match against a signature database (used bysecurity-scancommand andscan-suppressions.py)test— pass/fail against named specs (used by/nlpm:testandauditor-integration-test.yml)discover— produce a candidate list (/nlpm:lsdescription text andauditor-discover.yml)
Verbs proposed but not entered
Per the precedence order in analysis/vocabulary-design-principles.md (P6 last; warrant is an entry check, not a veto), proposed terms split into two lists.
Deferred (P2–P5 satisfied; awaiting warrant):
| Proposed verb | Closure gap or judgment named | Warrant needed |
|---|---|---|
triage | Finding-disposition (P4 closure: finding has no consumer verb) | User warrant — practitioner reaches for the term unprompted |
review (internal scope) | Human-in-the-loop reading, sanctioned cross-scope homonym with auditor review | User warrant — practitioner names a review act in internal artifacts |
rationalize | Vocabulary curation (P5: act of picking canonical from extracted frequencies has no name) | User warrant — practitioner names the curation act |
Deferred terms are documented but not yet canonical. R51 does not flag them as deprecated — they are nameless gaps, not synonyms.
Rejected by higher principle (no entry possible regardless of warrant):
| Proposed verb | Blocking principle | Reason |
|---|---|---|
audit (internal scope) | P1 | Merge test fails — score+check already cover internal-scope evaluation |
lint (internal scope) | P1 | Synonym of check; same identity criterion |
validate (internal scope) | P1 | Collapses to check in internal scope (canonical in auditor scope) |
analyze (internal scope) | P1 | Synonym of score; quantitative evaluation |
Nouns
Artifact-class nouns (rigid; survive state changes)
| Canonical noun | What it is | Examples |
|---|---|---|
artifact | any NL programming file Claude Code consumes | umbrella term |
command | commands/*.md, invoked via /<plugin>:<command> | commands/score.md |
agent | agents/*.md, invoked via Task tool with subagent_type | agents/scorer.md |
skill | skills/<plugin>/<name>/SKILL.md, auto-loaded by description match | skills/nlpm/rules/SKILL.md |
rule | a numbered item (R01–R50) inside the rules skill | inside skills/nlpm/rules/SKILL.md |
hook | shell script wired via hooks.json | hooks/hooks.json |
manifest | plugin.json or marketplace.json | .claude-plugin/plugin.json |
frontmatter | YAML block delimited by --- at the top of a Markdown file | every command, agent, skill |
Sanctioned homonyms (distinct senses, not drift — the drift scanner should not cluster these against the canonical nouns above):
| Term | Sanctioned sense | Where |
|---|---|---|
component | a structural part of the scored project's codebase — code modules, directories, services; may not be NL artifacts at all | R35/R49 rule text ("component map"), skills/nlpm/scoring/SKILL.md R35 row |
issue | a GitHub issue — the platform object the auditor opens, tracks, and closes | auditor scope: workflows, auditor-rule-review.yml, auditor pipeline prose in docs/auditor.md |
Everywhere else, component naming plugin pieces (command/agent/skill/hook)
and issue naming a detected problem are drift: the canonical nouns are
artifact and finding.
Implementation-role nouns (P2/P3 — agent names ride alongside command verbs)
| Role-noun | Paired verb | File |
|---|---|---|
scanner | ls | agents/scanner.md |
scorer | score | agents/scorer.md |
checker | check | agents/checker.md |
tester | test | agents/tester.md |
vague-scanner | score (sub-step) | agents/vague-scanner.md |
security-scanner | scan | agents/security-scanner.md |
These are not top-level verbs in their own right (P3). They are role-names for sub-step agents. Treat them as nouns naming the worker, not as operations on artifacts.
Show full SKILL.md (713 more words)Show less
Output-class nouns (produced by verbs)
| Noun | Produced by | Notes |
|---|---|---|
score | score verb | number 0–100 + penalty breakdown |
finding | score, check, scan | one problem detected; carries a fingerprint in auditor scope |
violation | check | a finding specifically from cross-reference checking |
penalty | score | the points subtracted by a single finding |
snapshot | score, trend | a point-in-time record appended to .claude/nlpm-history.json |
inventory | ls | the list of artifacts discovered in a path |
report | audit, report, trend | a roll-up document |
spec | test | a .nlpm-test/*.spec.md file defining expected behavior |
Auditor-scope nouns (only meaningful inside auditor/)
| Noun | What it is | File evidence |
|---|---|---|
case-study | post-merge article comparing original audit to HEAD | auditor/audits/<slug>.re-audit.md |
exemplar | teaching artifact from a high-scoring audit | auditor/exemplars/<slug>.md |
fingerprint | content-hash identity of a finding across re-runs | compute-fingerprint.sh |
event | one append-only record in events.jsonl | auditor/logs/events.jsonl |
disagreement | self/maintainer dispute over a finding | auditor/disagreements.jsonl |
registry | the tracked-repo database | auditor/registry/repos.json |
disposition | a finding's lifecycle status | enum values inside events.jsonl |
policy gate | a pre-PR check (no-external-PRs, CLA, pushback) | auditor-contribute.yml |
The bright-line table (P1, P5)
When the evaluation cluster overlaps in practice, use this table to disambiguate:
| Verb | Scope | Deterministic? | Judgment? | Output | Use when |
|---|---|---|---|---|---|
score | internal | yes | no | number + penalty list | quantifying quality with a rubric |
check | internal | yes | no | violation list | verifying cross-references and structural consistency |
test | both | yes | no | pass/fail | comparing actual behavior against a named spec |
scan | both | yes | no | findings against signatures | pattern-matching for a known class of problems |
audit | auditor | partly | yes | composite report | full quality assessment that combines score + scan + judgment |
review | auditor | no | yes (human) | comment trail | a human reads and forms an opinion |
Two verbs share a scope and an identity criterion ⇒ one must be retired or scope-split. Two verbs share an identity criterion across scopes ⇒ that is a sanctioned homonym (declare it in this table).
How to extend this registry
- Add a term only if it has warrant. The four warrant types come from
analysis/vocabulary-design-principles.mdP6: literary, user, structural, domain. Literary warrant is automatic via the extraction script — if the script picks the term up, it qualifies. - Re-run the extraction script.
python3 analysis/scripts/extract-vocabulary.pywritesanalysis/vocabulary-extract/summary.md. New terms appear there. - Slot the term. Add a row to the right table above (internal verb / auditor verb / artifact-class noun / role-noun / output-class noun / auditor-scope noun).
- Cite the file evidence. Every row has an "examples" or "file evidence" column. Cite at least one path.
- If retiring a synonym, list it under its canonical verb's "deprecated synonyms" line. Do not silently drop terms; deprecation is itself a vocabulary act and needs to be visible.
Deferred work (documented, not yet executed)
| Item | Why deferred | What unblocks it |
|---|---|---|
| Per-rule warrant tags on R01–R50 | Requires reading every rule + judging which warrant it earns its place by. User warrant data lives in rule-health.py outputs. | A separate auditor-refine-rules follow-up pass that combines rule-health data with this principle set. |
Adding triage as a /nlpm:triage command | P4 closure gap is real but triage has zero literary warrant today. User warrant is the missing evidence. | A practitioner reaching for the word unprompted in actual NLPM use. |
Resolving the agent-name-vs-command-name shadowing (scanner/ls, scorer/score) | Already documented above as a class boundary; no rename proposed. | A confirmed case of practitioner confusion. None observed today. |
| (resolved 2026-05-19) Workflow filename convention | Declared as sanctioned split. See "Auditor workflow filename convention" below. | — |
Auditor workflow filename convention
.github/workflows/auditor-*.yml filenames follow a two-rule split, both sanctioned:
| Pattern | Use when | Examples |
|---|---|---|
auditor-<verb> | The workflow changes state (gates a transition, opens a PR, classifies, moves a finding through its lifecycle) | auditor-audit.yml, auditor-contribute.yml, auditor-track.yml, auditor-classify.yml, auditor-refine-rules.yml, auditor-cite-exemplars.yml, auditor-discover.yml |
auditor-<output-noun> | The workflow's purpose is to produce a named artifact that downstream tooling consumes by name | auditor-case-study.yml, auditor-exemplar.yml, auditor-daily-report.yml, auditor-suppressions.yml, auditor-docs-diff.yml, auditor-rule-review.yml |
By P3, both patterns produce or gate, so both pass the top-level-verb test. The split is genuine: noun-named workflows are named after what they write, verb-named workflows after what they do.
Outlier: auditor-batch-processor.yml is named after a process role, not a verb or an output noun. Rename candidate: auditor-promote-batch. Deferred — git history, badge, and external references make the rename costly, and the existing name has literary warrant.
Scope note
This skill loads on requests to write, review, or name NLPM artifacts. For framework-level facts (hook event names, manifest field schemas, Claude Code naming conventions), see [[conventions]]. For the 50 quality rules (R01–R50), see [[rules]]. For penalty tables, see [[scoring]].
© xiaolai, ISC. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 1 other file in skills/nlpm/vocabulary of xiaolai/nlpm.
- SKILL.md
- registry.yaml
Open the folder on GitHubat commit bfa2fc2
Compare with similar skills
Vocabulary next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Vocabulary this skillxiaolai/nlpm | 150 | — | ~3.9k | Automated safety check: Pass | ISC | |
| MCP Server Builderanthropics/skills | 180k | 63 repos | ~2.3k | Automated safety check: Pass | Apache-2.0 | |
| Hook Development for Claude Code Pluginsanthropics/claude-plugins-official | 38k | 10 repos | ~4.1k | Automated safety check: Notes | Apache-2.0 | |
| Using Superpowersfarm-fe/farm | 5.6k | 36 repos | ~1.4k | Automated safety check: Pass | MIT | |
| Executing Plans Inlineobra/superpowers | 297k | 2 repos | ~5.1k | Automated safety check: Pass | MIT | |
| Skill CreatorAzure/azqr | 796 | 89 repos | ~8.2k | Automated safety check: Pass | Apache-2.0 |
Similar skills
- Official
- Official
Hook Development for Claude Code Plugins
anthropics/claude-plugins-official
Explains how to write Claude Code plugin hooks, both prompt-based checks and bash commands, for events such as PreToolUse, Stop and SessionStart.
Using Superpowers
farm-fe/farm
A skill your agent uses when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions
Executing Plans Inline
obra/superpowers
Has the agent carry out an implementation plan itself, task by task in the current session, keeping a ledger, proving each step with a test and ending with one whole-branch review.
- Official
Skill Creator
Azure/azqr
Create new skills, modify and improve existing skills, and measure skill performance.
- Official
Claude Code Agent Development
anthropics/claude-plugins-official
Explains how to write agents for Claude Code plugins: the markdown file with YAML frontmatter, trigger descriptions, model and color settings, and system prompt design.
More from xiaolai/nlpm
All 15 skills in this repoConventions
xiaolai/nlpm
Universal NL conventions: SKILL.md open spec, AGENTS.md, vague quantifiers, naming.
Conventions Antigravity
xiaolai/nlpm
Antigravity and Gemini CLI artifact schemas: .gemini/ paths, extensions, hooks.
Conventions Codex
xiaolai/nlpm
Codex CLI artifact schemas: config.toml, .codex-plugin, skills, hooks, AGENTS.md.
Orchestration
xiaolai/nlpm
Multi-agent workflow patterns: parallel dispatch, pipelines, QC gates, retries.
Patterns
xiaolai/nlpm
NL artifact anti-patterns: vague quantifiers, bare prohibitions, oversized skills.
Scoring
xiaolai/nlpm
100-point NL artifact rubric: penalty tables per artifact type, calibration cases.
Related
Categories
Questions about Vocabulary
What does Vocabulary do?
NLPM noun/verb registry (R51): pick the canonical term, detect vocabulary drift. Vocabulary is an agent skill from xiaolai/nlpm. NLPM noun/verb registry (R51): pick the canonical term, detect vocabulary drift.
When should I use Vocabulary?
Vocabulary fits situations like: agent Workflows work in your project.
How do I install Vocabulary in Claude Code?
Run `npx skills add xiaolai/nlpm --skill vocabulary -a claude-code`. Or copy the skill folder (skills/nlpm/vocabulary in xiaolai/nlpm) into .claude/skills/vocabulary in your project. Claude Code loads it when a task matches its description.
How do I install Vocabulary in Codex?
Run `npx skills add xiaolai/nlpm --skill vocabulary -a codex`. Or copy the skill folder (skills/nlpm/vocabulary in xiaolai/nlpm) into .agents/skills/vocabulary in your project. Codex loads it when a task matches its description.
Can I use Vocabulary in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add xiaolai/nlpm --skill vocabulary -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/vocabulary, .gemini/skills/vocabulary, .github/skills/vocabulary and .opencode/skills/vocabulary in your project.
What does Vocabulary need to run?
Going by SKILL.md and its folder, Vocabulary needs the command-line tools its instructions call (python3).
Does Vocabulary access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Vocabulary safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Vocabulary use?
Vocabulary is published under the ISC licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Vocabulary use?
About 3.9k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Vocabulary?
Skills that share tags, products or a category with Vocabulary: MCP Server Builder (anthropics/skills, 180k stars), Hook Development for Claude Code Plugins (anthropics/claude-plugins-official, 38k stars), Using Superpowers (farm-fe/farm, 5.6k stars) and Executing Plans Inline (obra/superpowers, 297k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Vocabulary?
xiaolai (a GitHub user) maintains it in xiaolai/nlpm, which has 150 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on October 10, 2026.
Source: xiaolai/nlpm on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.