siril9/presentation-skill / plugins/presentation-skill/skills/presentation-skill
presentation-skill
FlaggedBuild, edit, redesign, render, and verify content-led, structurally diverse, clean, editable, reproducible PowerPoint `.pptx` decks. Use for scientific, lab, clinical, consulting, board, investor, editorial, policy, and operational presentations where alignment, readability, contrast, and balanced density matter.
scripts/taste_grammar_catalog.py
View on GitHub#!/usr/bin/env python3
"""Canonical composition grammars and renderer role-system contracts."""
from __future__ import annotations
import copy
import json
from collections import Counter
from typing import Any
CATALOG_VERSION = "taste_grammar_catalog_v1"
RENDERER_ROLE_SYSTEM_VERSION = "renderer_role_systems_v1"
ROLE_SYSTEMS_VERSION = RENDERER_ROLE_SYSTEM_VERSION
ROLE_NAMES = (
"title",
"section",
"evidence",
"comparison",
"data",
"decision",
"references",
)
ROLE_ID_FIELDS = {role: f"{role}_system_id" for role in ROLE_NAMES}
SUPPORTED_ROLE_VARIANTS = {
"title",
"standard",
"split",
"cards-2",
"cards-3",
"timeline",
"stats",
"kpi-hero",
"table",
"lab-run-results",
"comparison-2col",
"matrix",
"flow",
"chart",
"image-sidebar",
"scientific-figure",
"generated-image",
}
COMPOSITION_GRAMMARS: dict[str, dict[str, Any]] = {
"consulting-answer-pyramid": {
"lane": "answer-led",
"description": "Lead with the answer, prove it with structured exhibits, and close on ownership.",
"style_presets": ["data-heavy-boardroom", "arctic-minimal"],
"prompt_keywords": ["board", "consulting", "executive", "recommendation", "strategy", "tradeoff"],
"role_system_ids": {
"title": "title-answer-ledger",
"section": "section-claim-chapters",
"evidence": "evidence-proof-stack",
"comparison": "comparison-option-scorecard",
"data": "data-executive-exhibit",
"decision": "decision-owner-commitment",
"references": "references-executive-notes",
},
"narrative_arc": {
"arc_id": "answer-context-proof-action",
"stages": ["answer", "context", "proof", "tradeoffs", "action"],
},
"density": {"level": "high", "distribution": "dense exhibits with short interpretive headlines"},
"grid": {"columns": 12, "gutter": "0.22-0.28in", "anchor": "left claim rail and aligned exhibit edge"},
"reading_path": ["headline answer", "primary exhibit", "implication and owner"],
"preferred_role_variants": {
"title": ["title"],
"section": ["standard", "kpi-hero"],
"evidence": ["chart", "table"],
"comparison": ["comparison-2col", "matrix"],
"data": ["chart", "table", "stats"],
"decision": ["table", "standard"],
"references": ["table"],
},
"invariant_moves": [
"State the governing answer before supporting detail.",
"Attach every exhibit to an explicit implication.",
"End with a named decision, owner, or next action.",
],
"forbidden_moves": [
"Do not delay the recommendation until the final slide.",
"Do not use decorative cards as a substitute for evidence hierarchy.",
],
},
"scientific-evidence-plate": {
"lane": "evidence-led",
"description": "Organize methods, figures, measurements, and limits as a traceable evidence plate.",
"style_presets": ["lab-report", "forest-research"],
"prompt_keywords": ["assay", "experiment", "figure", "lab", "method", "research", "results", "validation"],
"role_system_ids": {
"title": "title-study-plate",
"section": "section-method-result",
"evidence": "evidence-multipanel-plate",
"comparison": "comparison-control-cohort",
"data": "data-assay-readout",
"decision": "decision-evidence-threshold",
"references": "references-scientific-register",
},
"narrative_arc": {
"arc_id": "question-method-result-limit",
"stages": ["question", "method", "result", "interpretation", "limit", "next test"],
},
"density": {"level": "high", "distribution": "figure-dense center with compact methods and provenance rails"},
"grid": {"columns": 12, "gutter": "0.18-0.24in", "anchor": "figure plate with aligned captions and readouts"},
"reading_path": ["research question", "figure or assay plate", "interpretation", "limit and next test"],
"preferred_role_variants": {
"title": ["title"],
"section": ["standard"],
"evidence": ["scientific-figure", "image-sidebar"],
"comparison": ["comparison-2col", "table"],
"data": ["lab-run-results", "chart", "table"],
"decision": ["standard", "table"],
"references": ["table"],
},
"invariant_moves": [
"Keep methods, result, and interpretation visually distinguishable.",
"Carry units, denominators, and provenance beside the evidence.",
"State uncertainty or limitations before the recommendation.",
],
"forbidden_moves": [
"Do not crop or decorate evidence in ways that obscure interpretation.",
"Do not present a proxy metric as the final outcome without qualification.",
],
},
"clinical-care-pathway": {
"lane": "pathway-led",
"description": "Move from population and care state through evidence gates to a bounded clinical action.",
"style_presets": ["executive-clinical"],
"prompt_keywords": ["care", "clinical", "cohort", "diagnostic", "patient", "pathway", "safety", "treatment"],
"role_system_ids": {
"title": "title-clinical-status",
"section": "section-care-stage",
"evidence": "evidence-clinical-cohort",
"comparison": "comparison-care-options",
"data": "data-clinical-outcomes",
"decision": "decision-care-gate",
"references": "references-clinical-evidence",
},
"narrative_arc": {
"arc_id": "population-state-evidence-care-action",
"stages": ["population", "current state", "evidence gate", "care options", "action", "monitoring"],
},
"density": {"level": "medium-high", "distribution": "stage bands with one dominant clinical evidence object"},
"grid": {"columns": 12, "gutter": "0.22-0.28in", "anchor": "care-stage rail and outcome field"},
"reading_path": ["patient or cohort state", "evidence gate", "care decision", "monitoring implication"],
"preferred_role_variants": {
"title": ["title"],
"section": ["standard", "flow"],
"evidence": ["scientific-figure", "chart", "image-sidebar"],
"comparison": ["comparison-2col", "table"],
"data": ["chart", "table", "stats"],
"decision": ["flow", "table"],
"references": ["table"],
},
"invariant_moves": [
"Identify the patient or cohort state before discussing intervention.",
"Separate evidence threshold from care recommendation.",
"Make safety, monitoring, and escalation explicit.",
],
"forbidden_moves": [
"Do not imply clinical certainty beyond the cited evidence.",
"Do not hide contraindications or unresolved safety signals.",
],
},
"editorial-spread": {
"lane": "narrative-led",
"description": "Use editorial pacing, strong hierarchy, and selective evidence to build a readable argument.",
"style_presets": ["editorial-minimal", "paper-journal"],
"prompt_keywords": ["article", "editorial", "essay", "feature", "journal", "narrative", "public", "story"],
"role_system_ids": {
"title": "title-editorial-masthead",
"section": "section-editorial-folio",
"evidence": "evidence-captioned-feature",
"comparison": "comparison-editorial-columns",
"data": "data-annotated-graphic",
"decision": "decision-editorial-takeaway",
"references": "references-editorial-notes",
},
"narrative_arc": {
"arc_id": "premise-scene-evidence-turn-close",
"stages": ["premise", "scene", "evidence", "turn", "implication", "close"],
},
"density": {"level": "medium", "distribution": "alternating open spreads and evidence-rich features"},
"grid": {"columns": 10, "gutter": "0.24-0.32in", "anchor": "masthead baseline and asymmetric feature column"},
"reading_path": ["editorial headline", "feature object", "caption or annotation", "closing implication"],
"preferred_role_variants": {
"title": ["title"],
"section": ["standard", "generated-image"],
"evidence": ["image-sidebar", "generated-image", "scientific-figure"],
"comparison": ["split", "comparison-2col"],
"data": ["chart", "table"],
"decision": ["standard"],
"references": ["table"],
},
"invariant_moves": [
"Alternate open narrative space with denser proof moments.",
"Use captions and annotations to connect visuals to the argument.",
"Keep one clear editorial turn per section.",
],
"forbidden_moves": [
"Do not repeat the same card grid across consecutive slides.",
"Do not use decorative imagery without an explanatory role.",
],
},
"investor-thesis-stage": {
"lane": "thesis-led",
"description": "Stage the market thesis, proof, economics, risk, and ask as a cumulative investment case.",
"style_presets": ["sunset-investor", "bold-startup-narrative"],
"prompt_keywords": ["ask", "fundraising", "growth", "investor", "market", "pitch", "revenue", "startup"],
"role_system_ids": {
"title": "title-investor-thesis",
"section": "section-thesis-stage",
"evidence": "evidence-market-proof",
"comparison": "comparison-market-position",
"data": "data-unit-economics",
"decision": "decision-investment-ask",
"references": "references-investor-diligence",
},
"narrative_arc": {
"arc_id": "problem-window-proof-economics-ask",
"stages": ["problem", "market window", "solution", "proof", "economics", "risk", "ask"],
},
"density": {"level": "medium", "distribution": "high-contrast thesis stages punctuated by dense proof slides"},
"grid": {"columns": 12, "gutter": "0.24-0.30in", "anchor": "thesis field with proof and downside rails"},
"reading_path": ["thesis", "proof signal", "economic implication", "risk", "ask"],
"preferred_role_variants": {
"title": ["title"],
"section": ["kpi-hero", "standard"],
"evidence": ["stats", "chart", "generated-image"],
"comparison": ["comparison-2col", "matrix"],
"data": ["chart", "stats", "table"],
"decision": ["kpi-hero", "standard"],
"references": ["table"],
},
"invariant_moves": [
"Tie each proof point to the investment thesis.",
"Show the economic mechanism, not only market size.",
"State the ask, milestone, and principal downside clearly.",
],
"forbidden_moves": [
"Do not substitute vanity metrics for durable proof.",
"Do not present upside without a bounded risk or assumption.",
],
},
"operations-grid": {
"lane": "state-led",
"description": "Expose operating state, variance, dependencies, owners, and the next control action.",
"style_presets": ["lavender-ops", "charcoal-safety"],
"prompt_keywords": ["incident", "operations", "owner", "process", "risk", "safety", "status", "workflow"],
"role_system_ids": {
"title": "title-operations-state",
"section": "section-operating-cycle",
"evidence": "evidence-operations-signal",
"comparison": "comparison-plan-actual",
"data": "data-operations-grid",
"decision": "decision-control-action",
"references": "references-operations-log",
},
"narrative_arc": {
"arc_id": "state-variance-cause-control-owner",
"stages": ["state", "variance", "cause", "control", "owner", "follow-up"],
},
"density": {"level": "high", "distribution": "stable operating grid with bounded exception emphasis"},
"grid": {"columns": 12, "gutter": "0.18-0.24in", "anchor": "status index, variance field, and owner column"},
"reading_path": ["operating state", "exception or variance", "root cause", "control and owner"],
"preferred_role_variants": {
"title": ["title"],
"section": ["stats", "standard"],
"evidence": ["chart", "flow", "table"],
"comparison": ["comparison-2col", "table"],
"data": ["stats", "table", "chart"],
"decision": ["table", "flow"],
"references": ["table"],
},
"invariant_moves": [
"Separate baseline state from exceptions and escalation.",
"Connect every variance to a cause, control, or owner.",
"Keep repeated operational coordinates stable across slides.",
],
"forbidden_moves": [
"Do not use color as the only status encoding.",
"Do not hide unresolved exceptions inside aggregate metrics.",
],
},
"policy-public-docket": {
"lane": "docket-led",
"description": "Frame the public question, affected groups, evidence, options, implementation, and accountability.",
"style_presets": ["warm-terracotta"],
"prompt_keywords": ["civic", "community", "government", "implementation", "policy", "public", "stakeholder"],
"role_system_ids": {
"title": "title-public-question",
"section": "section-policy-docket",
"evidence": "evidence-public-record",
"comparison": "comparison-policy-options",
"data": "data-public-impact",
"decision": "decision-policy-recommendation",
"references": "references-public-docket",
},
"narrative_arc": {
"arc_id": "public-question-impact-options-implementation",
"stages": ["public question", "affected groups", "evidence", "options", "recommendation", "implementation", "accountability"],
},
"density": {"level": "medium-high", "distribution": "open public framing followed by evidence and option registers"},
"grid": {"columns": 12, "gutter": "0.24-0.30in", "anchor": "question header with stakeholder and evidence columns"},
"reading_path": ["public question", "affected groups", "evidence", "options", "implementation and accountability"],
"preferred_role_variants": {
"title": ["title"],
"section": ["standard", "timeline"],
"evidence": ["image-sidebar", "chart", "table"],
"comparison": ["matrix", "comparison-2col"],
"data": ["chart", "table"],
"decision": ["table", "timeline"],
"references": ["table"],
},
"invariant_moves": [
"Name affected groups before comparing policy options.",
"Separate evidence, value judgments, and implementation assumptions.",
"Close with accountability, timing, and measurable public outcomes.",
],
"forbidden_moves": [
"Do not collapse stakeholder impacts into one aggregate score.",
"Do not state implementation certainty without owners and constraints.",
],
},
"technical-telemetry-canvas": {
"lane": "telemetry-led",
"description": "Trace system state from architecture and signals through failure modes to remediation.",
"style_presets": ["midnight-neon"],
"prompt_keywords": ["architecture", "engineering", "incident", "latency", "platform", "system", "technical", "telemetry"],
"role_system_ids": {
"title": "title-telemetry-state",
"section": "section-system-layer",
"evidence": "evidence-signal-trace",
"comparison": "comparison-architecture-tradeoff",
"data": "data-telemetry-canvas",
"decision": "decision-remediation-gate",
"references": "references-technical-register",
},
"narrative_arc": {
"arc_id": "system-state-signal-failure-remediation",
"stages": ["system scope", "architecture", "signal", "failure mode", "remediation", "verification"],
},
"density": {"level": "high", "distribution": "signal-dense canvas with explicit diagnostic and action zones"},
"grid": {"columns": 12, "gutter": "0.18-0.24in", "anchor": "system layer rail, telemetry field, and remediation lane"},
"reading_path": ["system state", "signal trace", "failure mode", "remediation", "verification"],
"preferred_role_variants": {
"title": ["title"],
"section": ["flow", "standard"],
"evidence": ["flow", "chart", "image-sidebar"],
"comparison": ["comparison-2col", "matrix"],
"data": ["chart", "stats", "table"],
"decision": ["flow", "table"],
"references": ["table"],
},
"invariant_moves": [
"Anchor every signal to a named system layer or component.",
"Distinguish observation, diagnosis, remediation, and verification.",
"Keep units, thresholds, and time windows visible.",
],
"forbidden_moves": [
"Do not present telemetry without a baseline or threshold.",
"Do not conflate correlation with a confirmed failure cause.",
],
},
}
PRESET_TO_GRAMMAR = {
preset: grammar_id
for grammar_id, grammar in COMPOSITION_GRAMMARS.items()
for preset in grammar["style_presets"]
}
def _role_system_catalogs() -> dict[str, list[dict[str, str]]]:
catalogs: dict[str, list[dict[str, str]]] = {}
for role in ROLE_NAMES:
seen: dict[str, dict[str, str]] = {}
for grammar_id, grammar in COMPOSITION_GRAMMARS.items():
system_id = str(grammar["role_system_ids"][role])
seen[system_id] = {
"system_id": system_id,
"role": role,
"composition_grammar_id": grammar_id,
"intent": str(grammar["description"]),
}
catalogs[f"{role}_systems"] = list(seen.values())
return catalogs
ROLE_SYSTEM_CATALOGS = _role_system_catalogs()
def composition_grammar(grammar_id: str) -> dict[str, Any]:
grammar = COMPOSITION_GRAMMARS.get(str(grammar_id or "").strip())
if grammar is None:
raise KeyError(f"unknown composition grammar: {grammar_id!r}")
return copy.deepcopy({"grammar_id": grammar_id, **grammar})
def grammar_for_preset(preset: str) -> dict[str, Any]:
key = str(preset or "").strip() or "executive-clinical"
grammar_id = PRESET_TO_GRAMMAR.get(key, PRESET_TO_GRAMMAR["executive-clinical"])
return composition_grammar(grammar_id)
def renderer_role_systems_for_grammar(grammar_id: str, *, preset: str = "") -> dict[str, Any]:
grammar = composition_grammar(grammar_id)
role_ids = grammar["role_system_ids"]
payload: dict[str, Any] = {
"schema_version": RENDERER_ROLE_SYSTEM_VERSION,
"catalog_version": CATALOG_VERSION,
"composition_grammar_id": grammar_id,
"style_preset": str(preset or "").strip() or grammar["style_presets"][0],
"narrative_arc": grammar["narrative_arc"],
"density": grammar["density"],
"grid": grammar["grid"],
"reading_path": grammar["reading_path"],
"preferred_role_variants": grammar["preferred_role_variants"],
"invariant_moves": grammar["invariant_moves"],
"forbidden_moves": grammar["forbidden_moves"],
}
for role, field in ROLE_ID_FIELDS.items():
payload[field] = role_ids[role]
return copy.deepcopy(payload)
def renderer_role_systems_for_preset(preset: str) -> dict[str, Any]:
key = str(preset or "").strip() or "executive-clinical"
grammar_id = PRESET_TO_GRAMMAR.get(key, PRESET_TO_GRAMMAR["executive-clinical"])
resolved_preset = key if key in PRESET_TO_GRAMMAR else "executive-clinical"
return renderer_role_systems_for_grammar(grammar_id, preset=resolved_preset)
def build_taste_grammar_catalog() -> dict[str, Any]:
grammars = [composition_grammar(grammar_id) for grammar_id in COMPOSITION_GRAMMARS]
narrative_arcs = [copy.deepcopy(grammar["narrative_arc"]) for grammar in grammars]
return {
"catalog_version": CATALOG_VERSION,
"renderer_role_systems_version": RENDERER_ROLE_SYSTEM_VERSION,
"grammar_count": len(grammars),
"preset_count": len(PRESET_TO_GRAMMAR),
"grammars": grammars,
"preset_to_grammar": dict(PRESET_TO_GRAMMAR),
"role_system_catalogs": copy.deepcopy(ROLE_SYSTEM_CATALOGS),
"narrative_arcs": narrative_arcs,
}
def validate_renderer_role_systems_v1(
payload: Any,
*,
expected_preset: str = "",
) -> list[str]:
failures: list[str] = []
if not isinstance(payload, dict):
return ["must be an object"]
if payload.get("schema_version") != RENDERER_ROLE_SYSTEM_VERSION:
failures.append(f"schema_version must be {RENDERER_ROLE_SYSTEM_VERSION!r}")
if payload.get("catalog_version") != CATALOG_VERSION:
failures.append(f"catalog_version must be {CATALOG_VERSION!r}")
grammar_id = str(payload.get("composition_grammar_id") or "").strip()
grammar = COMPOSITION_GRAMMARS.get(grammar_id)
if grammar is None:
failures.append(f"unknown composition_grammar_id {grammar_id!r}")
return failures
preset = str(payload.get("style_preset") or "").strip()
if preset not in grammar["style_presets"]:
failures.append(f"style_preset {preset!r} is not mapped to {grammar_id!r}")
if expected_preset and preset != expected_preset:
failures.append(f"style_preset {preset!r} does not match expected preset {expected_preset!r}")
for role, field in ROLE_ID_FIELDS.items():
value = payload.get(field)
if not isinstance(value, str) or not value.strip():
failures.append(f"{field} must be a non-empty string")
elif value != grammar["role_system_ids"][role]:
failures.append(f"{field} {value!r} does not match grammar {grammar_id!r}")
narrative_arc = payload.get("narrative_arc")
if not isinstance(narrative_arc, dict):
failures.append("narrative_arc must be an object")
else:
if narrative_arc.get("arc_id") != grammar["narrative_arc"]["arc_id"]:
failures.append("narrative_arc.arc_id does not match the composition grammar")
if not isinstance(narrative_arc.get("stages"), list) or len(narrative_arc["stages"]) < 3:
failures.append("narrative_arc.stages must contain at least three stages")
for key in ("density", "grid"):
if not isinstance(payload.get(key), dict) or not payload[key]:
failures.append(f"{key} must be a non-empty object")
reading_path = payload.get("reading_path")
if not isinstance(reading_path, list) or len(reading_path) < 2 or not all(
isinstance(item, str) and item.strip() for item in reading_path
):
failures.append("reading_path must contain at least two non-empty strings")
variants = payload.get("preferred_role_variants")
if not isinstance(variants, dict):
failures.append("preferred_role_variants must be an object")
else:
for role in ROLE_NAMES:
values = variants.get(role)
if not isinstance(values, list) or not values:
failures.append(f"preferred_role_variants.{role} must be a non-empty list")
continue
invalid = [value for value in values if value not in SUPPORTED_ROLE_VARIANTS]
if invalid:
failures.append(f"preferred_role_variants.{role} contains unsupported values {invalid}")
for key in ("invariant_moves", "forbidden_moves"):
values = payload.get(key)
if not isinstance(values, list) or not values or not all(
isinstance(item, str) and item.strip() for item in values
):
failures.append(f"{key} must be a non-empty string list")
return failures
def validate_taste_grammar_catalog(catalog: dict[str, Any] | None = None) -> dict[str, Any]:
payload = catalog or build_taste_grammar_catalog()
grammars = payload.get("grammars") if isinstance(payload.get("grammars"), list) else []
failures: list[str] = []
grammar_ids = [str(grammar.get("grammar_id") or "") for grammar in grammars if isinstance(grammar, dict)]
expected_ids = list(COMPOSITION_GRAMMARS)
if len(grammars) != 8:
failures.append(f"grammar_count={len(grammars)} expected=8")
if set(grammar_ids) != set(expected_ids):
failures.append("composition grammar IDs do not match the canonical eight")
preset_map = payload.get("preset_to_grammar") if isinstance(payload.get("preset_to_grammar"), dict) else {}
if len(preset_map) != 13:
failures.append(f"preset_count={len(preset_map)} expected=13")
usage = Counter(map(str, preset_map.values()))
overused = {grammar_id: count for grammar_id, count in usage.items() if count > 2}
if overused:
failures.append(f"composition grammars mapped to more than two presets: {overused}")
catalogs = payload.get("role_system_catalogs") if isinstance(payload.get("role_system_catalogs"), dict) else {}
counts = {
role: len(catalogs.get(f"{role}_systems") or [])
for role in ROLE_NAMES
}
if counts["title"] != 8:
failures.append(f"title_system_count={counts['title']} expected=8")
if counts["section"] < 6:
failures.append(f"section_system_count={counts['section']} expected>=6")
if counts["evidence"] < 8:
failures.append(f"evidence_system_count={counts['evidence']} expected>=8")
if counts["data"] < 8:
failures.append(f"data_system_count={counts['data']} expected>=8")
narrative_arcs = payload.get("narrative_arcs") if isinstance(payload.get("narrative_arcs"), list) else []
arc_ids = {
str(arc.get("arc_id") or "")
for arc in narrative_arcs
if isinstance(arc, dict) and str(arc.get("arc_id") or "").strip()
}
if len(arc_ids) < 6:
failures.append(f"narrative_arc_count={len(arc_ids)} expected>=6")
for preset in preset_map:
role_failures = validate_renderer_role_systems_v1(
renderer_role_systems_for_preset(preset),
expected_preset=preset,
)
failures.extend(f"{preset}: {failure}" for failure in role_failures)
return {
"passed": not failures,
"catalog_version": payload.get("catalog_version"),
"renderer_role_systems_version": payload.get("renderer_role_systems_version"),
"grammar_count": len(grammars),
"preset_count": len(preset_map),
"max_presets_per_grammar": max(usage.values(), default=0),
"role_system_counts": counts,
"narrative_arc_count": len(arc_ids),
"failures": failures,
}
def main() -> int:
print(json.dumps(validate_taste_grammar_catalog(), indent=2, ensure_ascii=False))
return 0
if __name__ == "__main__":
raise SystemExit(main())
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |