--- name: "whiteboard-to-infographic" description: "Convert a hand-drawn whiteboard photo (or sketched process/diagram) from a discovery or working session into a polished, client-ready infographic as an editable PowerPoint slide (.pptx). Use this whenever someone uploads or references a photo of a whiteboard, a hand sketch of a process or system, or session-capture notes and wants it \"cleaned up\", \"made presentable\", \"turned into a diagram/infographic\", \"converted for the client\", or similar \u2014 even if they don't say the word \"infographic\". Covers two layout archetypes \u2014 linear process flows (left-to-right step rails) and network/topology maps (locations, systems, or actors connected by labeled flows). Built for consulting/ERP discovery capture but works for any whiteboard-to-deliverable conversion. Do NOT use for charts/graphs from numeric data, for editing an existing polished design, or for generating diagrams with no source sketch." --- # Whiteboard to Infographic Turn a messy hand-drawn capture into a clean, branded one-page infographic — delivered as a single **editable PowerPoint slide** — that a consultant can put in front of a client. The hard, non-delegable work is reading the handwriting correctly and choosing the right structure; the styling and layout are codified here so every output looks consistent. ## The discipline (do this in order — do not skip ahead to styling) This is structure-first. Producing a polished deliverable from a misread sketch is the main failure mode, so confirm interpretation **before** building. 1. **Read the source carefully.** The photo is the source of truth. Zoom and crop aggressively — examine the image, then crop it into regions and re-examine at 2–3x so the handwriting is legible. Transcribe every box, label, arrow, annotation, and margin note. Do not silently drop anything you can't read — list it as a question. 2. **Confirm interpretation and lock structure.** Before building the slide, send the user a compact transcription: the nodes/steps, the connections/flows, and any side notes or requirements. Explicitly flag: ambiguous words, unexpanded acronyms, and anything you're inferring. Ask the questions that change the build (see "Questions to ask"). Wait for confirmation. Build only what's on the board — never invent steps, fields, or connections to make it look complete. If a relationship isn't drawn, don't draw it. 3. **Get the color palette.** The palette is client-specific and is NOT baked into this skill. If the user gave colors in the prompt or context, use them. Otherwise, ask before building. You need hex values for these roles: - **ink** — primary structure, dark text, node borders (e.g. a deep navy) - **secondary** — sub-heads, connectors, region boundaries (e.g. steel blue) - **accent** — sequence numbers, bullets, flags, emphasis (e.g. a signal red) - **card** — light fill for callout cards / cluster fills (e.g. pale blue-gray) - **accent-tint** — light wash behind accent-flagged items (e.g. pale red) Derive sensible text tints (e.g. a muted slate for sub-captions) from these. Offer to suggest a palette if the user has no preference, but don't assume. 4. **Choose the archetype** (state your choice and why): - **Process rail** — the sketch is a sequence/flow with a clear order (step 1 → step 2 → …). Read `references/process-rail.md`. - **Network map** — the sketch is locations/systems/actors connected by flows, with no single linear order (a topology). Read `references/network-map.md`. If it's genuinely both, lead with the dominant reading and ask. 5. **Build, verify, iterate.** Build the slide with `python-pptx` per the chosen archetype's reference. The layout is deterministic — **code decides every coordinate** — so verify structurally rather than by eye: every step/node present, columns/nodes evenly spaced, no shape overlapping another, every label inside its box, drop-lines landing on their cards, edges approaching nodes straight-on and not crossing through them. If your runtime can rasterize the slide, optionally do a visual critique pass; if it can't, lean entirely on the computed geometry. Fix in code and rebuild. Faithfulness beats polish — never add a step, field, or connection that wasn't on the board. ## Building the slide (native python-pptx) Build the infographic as a **single editable slide** with `python-pptx`, drawn natively as shapes — no HTML, no image rendering, no external render step. It runs inside the agent's Python container and returns a `.pptx` the client can open and edit. The layout is **deterministic — code decides every coordinate.** Design on a fixed 1600×1000 grid mapped to a 13.333in slide (see each reference for the `PX()` scale helper), and compute every position from the step/node count and the canvas so the slide is correct by construction. Use the house-style system fonts (`Arial Narrow` for condensed display, `Arial` for body) so nothing depends on font downloads. Turn off the default autoshape shadow (`shape.shadow.inherit = False`) for the flat house look. Save the deck and return the single `.pptx` as the deliverable — no PNG/PDF/HTML side-artifacts. **Revisions: never reuse a filename.** When the user asks for changes to an infographic you have already delivered, build the revision as a **new file with a new name** — `infographic-v1.pptx`, `infographic-v2.pptx`, `infographic-v3.pptx`, and so on — incrementing on every subsequent revision for the life of the conversation. Do not overwrite, re-save, or re-deliver an existing filename: a reused name can prevent the updated file from reaching the user, who then receives the earlier version or no file at all. One revision, one new filename, every time. ## Conventions (shared by both archetypes) These are the house style. Keep them consistent across a client engagement. - **Header band**: condensed uppercase title + a thin small-caps subtitle, a full-width accent rule under it, and an optional right-aligned scope/context note. A short eyebrow label (e.g. "THE PROCESS", "SUPPLY NETWORK") sits above each major band. - **Accent flags** encode meaning — don't decorate with them: - A solid **accent "GAP" pill** marks a gap between the standard system and the client's process (a fit/gap item). Use only where the user confirms a gap. - A hollow **"need details" tag** marks an open item to define in follow-up. - Include a one-line legend explaining any flag you use. - **Faithfulness**: behaviors/notes with no home in the visual go in a list, not invented shapes. Keep the client's exact terms and acronyms; don't expand or rename acronyms the client already knows unless asked. - **Restraint**: spend boldness on one signature device (the numbered chevron rail, or the region cluster), keep everything else quiet. Whitespace where the source is sparse is honest — don't pad it with invented detail. ## Questions to ask before building Ask only the ones that actually change the build; don't interrogate. - Unreadable words / unexpanded acronyms (transcribe what you can, ask the rest). - Archetype if ambiguous (rail vs. map). - For a rail: is there a meaningful sequence number/priority on any step? What goes in the bottom band — principles, requirements, or nothing? - For a map: which connections must be explicit vs. summarized in a legend; what each node/edge label means; does anything apply across all nodes. - Whether a "detailed" companion version is wanted (dense capture of every margin note) in addition to a clean executive version. - The color palette, if not already provided. ## References - `references/process-rail.md` — linear flow archetype: deterministic banded layout (header, numbered chevron rail with drop-lines, callout cards, principles/requirements bottom band), as a `python-pptx` skeleton with color placeholders. - `references/network-map.md` — topology archetype: node/edge/cluster helpers (labeled edges with arrowheads, dashed region clusters, edge-label chips, flow legend, requirements band), as a `python-pptx` skeleton with color placeholders. ## Run this — do not improvise This capability's deterministic implementation is a RAPP single-file agent, linked beside this file as `whiteboard_to_infographic_agent.py` and embedded as the fenced Python below (sha256 caa451059d974d14…; a byte-exact copy is also vaulted in the capsule comment at the end of this file). On a host with sandbox execution, run the linked file directly — if it is missing, write the fence contents verbatim to `whiteboard_to_infographic_agent.py` first: ```bash python3 whiteboard_to_infographic_agent.py '{"key": "value"}' # arguments as one JSON object echo '{"key": "value"}' | python3 whiteboard_to_infographic_agent.py # or on stdin python3 whiteboard_to_infographic_agent.py --tool # emit the JSON tool contract ``` Treat stdout as a tool result. If it reports missing or unresolved inputs, stop and collect them. If it returns `steps`, execute those steps in order exactly as returned; if it returns `instructions`, follow them with the supplied inputs. Otherwise use the result verbatim. Do not invent behavior beyond that output. On a host without code execution, treat the Parameters schema and the code below as the exact specification and never paraphrase a step. Never edit inside the generated markers; a converter-equipped host can instead restore the original file checksum-verified with the installed `rapp-agent-converter/scripts/toast.py convert SKILL.md --to agent`. ```python # rapp:deterministic """WhiteboardToInfographic -- Convert a hand-drawn whiteboard photo (or sketched process/diagram) from a discovery or working session into a polished, client-ready infographic as an editable PowerPoint slide (.pptx). Use this whenever someone uploads or references a photo of a whiteboard, a hand sketch of a process or system, or session-capture notes and wants it "cleaned up", "made presentable", "turned into a diagram/infographic", "converted for the client", or similar — even if they don't say the word "infographic". Covers two layout archetypes — linear process flows (left-to-right step rails) and network/topology maps (locations, systems, or actors connected by labeled flows). Built for consulting/ERP discovery capture but works for any whiteboard-to-deliverable conversion. Do NOT use for charts/graphs from numeric data, for editing an existing polished design, or for generating diagrams with no source sketch. Generated by the rapp skill from whiteboard-to-infographic. The RCI capsule at the bottom of this file carries the full original; `toast.py convert` restores it byte-exact.""" import json import re import sys try: from agents.basic_agent import BasicAgent except ImportError: # running OUTSIDE a brainstem -- stay executable anyway. class BasicAgent: # noqa: D101 - minimal stand-in, same contract def __init__(self, name=None, metadata=None): if name: self.name = name if metadata: self.metadata = metadata def perform(self, **kwargs): return "Not implemented." def system_context(self): return None def to_tool(self): return {"type": "function", "function": { "name": self.name, "description": self.metadata.get("description", ""), "parameters": self.metadata.get("parameters", {})}} # The procedural layer, verbatim from the source capability. INSTRUCTIONS = '# Whiteboard to Infographic\n\nTurn a messy hand-drawn capture into a clean, branded one-page infographic —\ndelivered as a single **editable PowerPoint slide** — that a consultant can put\nin front of a client. The hard, non-delegable work is reading the handwriting\ncorrectly and choosing the right structure; the styling and layout are codified\nhere so every output looks consistent.\n\n## The discipline (do this in order — do not skip ahead to styling)\n\nThis is structure-first. Producing a polished deliverable from a misread sketch\nis the main failure mode, so confirm interpretation **before** building.\n\n1. **Read the source carefully.** The photo is the source of truth. Zoom and\n crop aggressively — examine the image, then crop it into regions and\n re-examine at 2–3x so the handwriting is legible. Transcribe every box,\n label, arrow, annotation, and margin note. Do not silently drop anything you\n can\'t read — list it as a question.\n\n2. **Confirm interpretation and lock structure.** Before building the slide,\n send the user a compact transcription: the nodes/steps, the\n connections/flows, and any side notes or requirements. Explicitly flag:\n ambiguous words, unexpanded acronyms, and anything you\'re inferring. Ask the\n questions that change the build (see "Questions to ask"). Wait for\n confirmation. Build only what\'s on the board — never invent steps, fields, or\n connections to make it look complete. If a relationship isn\'t drawn, don\'t\n draw it.\n\n3. **Get the color palette.** The palette is client-specific and is NOT baked\n into this skill. If the user gave colors in the prompt or context, use them.\n Otherwise, ask before building. You need hex values for these roles:\n - **ink** — primary structure, dark text, node borders (e.g. a deep navy)\n - **secondary** — sub-heads, connectors, region boundaries (e.g. steel blue)\n - **accent** — sequence numbers, bullets, flags, emphasis (e.g. a signal red)\n - **card** — light fill for callout cards / cluster fills (e.g. pale blue-gray)\n - **accent-tint** — light wash behind accent-flagged items (e.g. pale red)\n Derive sensible text tints (e.g. a muted slate for sub-captions) from these.\n Offer to suggest a palette if the user has no preference, but don\'t assume.\n\n4. **Choose the archetype** (state your choice and why):\n - **Process rail** — the sketch is a sequence/flow with a clear order\n (step 1 → step 2 → …). Read `references/process-rail.md`.\n - **Network map** — the sketch is locations/systems/actors connected by\n flows, with no single linear order (a topology). Read\n `references/network-map.md`.\n If it\'s genuinely both, lead with the dominant reading and ask.\n\n5. **Build, verify, iterate.** Build the slide with `python-pptx` per the chosen\n archetype\'s reference. The layout is deterministic — **code decides every\n coordinate** — so verify structurally rather than by eye: every step/node\n present, columns/nodes evenly spaced, no shape overlapping another, every\n label inside its box, drop-lines landing on their cards, edges approaching\n nodes straight-on and not crossing through them. If your runtime can\n rasterize the slide, optionally do a visual critique pass; if it can\'t, lean\n entirely on the computed geometry. Fix in code and rebuild. Faithfulness \n beats polish — never add a step, field, or connection that wasn\'t on the \n board.\n\n## Building the slide (native python-pptx)\n\nBuild the infographic as a **single editable slide** with `python-pptx`, drawn\nnatively as shapes — no HTML, no image rendering, no external render step. It\nruns inside the agent\'s Python container and returns a `.pptx` the client can\nopen and edit.\n\nThe layout is **deterministic — code decides every coordinate.** Design on a\nfixed 1600×1000 grid mapped to a 13.333in slide (see each reference for the\n`PX()` scale helper), and compute every position from the step/node count and\nthe canvas so the slide is correct by construction. Use the house-style system\nfonts (`Arial Narrow` for condensed display, `Arial` for body) so nothing\ndepends on font downloads. Turn off the default autoshape shadow\n(`shape.shadow.inherit = False`) for the flat house look. Save the deck and\nreturn the single `.pptx` as the deliverable — no PNG/PDF/HTML side-artifacts.\n\n**Revisions: never reuse a filename.** When the user asks for changes to an\ninfographic you have already delivered, build the revision as a **new file with\na new name** — `infographic-v1.pptx`, `infographic-v2.pptx`, `infographic-v3.pptx`,\nand so on — incrementing on every subsequent revision for the life of the\nconversation. Do not overwrite, re-save, or re-deliver an existing filename: a\nreused name can prevent the updated file from reaching the user, who then\nreceives the earlier version or no file at all. One revision, one new filename,\nevery time.\n\n## Conventions (shared by both archetypes)\n\nThese are the house style. Keep them consistent across a client engagement.\n\n- **Header band**: condensed uppercase title + a thin small-caps subtitle, a\n full-width accent rule under it, and an optional right-aligned scope/context\n note. A short eyebrow label (e.g. "THE PROCESS", "SUPPLY NETWORK") sits above\n each major band.\n- **Accent flags** encode meaning — don\'t decorate with them:\n - A solid **accent "GAP" pill** marks a gap between the standard system and the\n client\'s process (a fit/gap item). Use only where the user confirms a gap.\n - A hollow **"need details" tag** marks an open item to define in follow-up.\n - Include a one-line legend explaining any flag you use.\n- **Faithfulness**: behaviors/notes with no home in the visual go in a list, not\n invented shapes. Keep the client\'s exact terms and acronyms; don\'t expand or\n rename acronyms the client already knows unless asked.\n- **Restraint**: spend boldness on one signature device (the numbered chevron\n rail, or the region cluster), keep everything else quiet. Whitespace where the\n source is sparse is honest — don\'t pad it with invented detail.\n\n## Questions to ask before building\n\nAsk only the ones that actually change the build; don\'t interrogate.\n- Unreadable words / unexpanded acronyms (transcribe what you can, ask the rest).\n- Archetype if ambiguous (rail vs. map).\n- For a rail: is there a meaningful sequence number/priority on any step? What\n goes in the bottom band — principles, requirements, or nothing?\n- For a map: which connections must be explicit vs. summarized in a legend;\n what each node/edge label means; does anything apply across all nodes.\n- Whether a "detailed" companion version is wanted (dense capture of every\n margin note) in addition to a clean executive version.\n- The color palette, if not already provided.\n\n## References\n\n- `references/process-rail.md` — linear flow archetype: deterministic banded\n layout (header, numbered chevron rail with drop-lines, callout cards,\n principles/requirements bottom band), as a `python-pptx` skeleton with color\n placeholders.\n- `references/network-map.md` — topology archetype: node/edge/cluster helpers\n (labeled edges with arrowheads, dashed region clusters, edge-label chips, flow\n legend, requirements band), as a `python-pptx` skeleton with color\n placeholders.' # Ordered commands lifted verbatim from the capability's own documentation. STEPS = [] class WhiteboardToInfographicAgent(BasicAgent): def __init__(self): self.name = 'WhiteboardToInfographic' self.metadata = { "name": "WhiteboardToInfographic", "description": "Convert a hand-drawn whiteboard photo (or sketched process/diagram) from a discovery or working session into a polished, client-ready infographic as an editable PowerPoint slide (.pptx). Use this whenever someone uploads or references a photo of a whiteboard, a hand sketch of a process or system, or session-capture notes and wants it \"cleaned up\", \"made presentable\", \"turned into a diagram/infographic\", \"converted for the client\", or similar \u2014 even if they don't say the word \"infographic\". Covers two layout archetypes \u2014 linear process flows (left-to-right step rails) and network/topology maps (locations, systems, or actors connected by labeled flows). Built for consulting/ERP discovery capture but works for any whiteboard-to-deliverable conversion. Do NOT use for charts/graphs from numeric data, for editing an existing polished design, or for generating diagrams with no source sketch.", "parameters": { "type": "object", "properties": {}, "required": [] } } super().__init__(name=self.name, metadata=self.metadata) def perform(self, **kwargs): # toaster:generated-perform return json.dumps({"status": "ok", "instructions": INSTRUCTIONS, "inputs": kwargs, "note": "Prose-only capability: follow INSTRUCTIONS " "with the given inputs."}, indent=2) if __name__ == "__main__": # echo '{"arg": "value"}' | python3 whiteboard_to_infographic_agent.py # python3 whiteboard_to_infographic_agent.py '{"arg": "value"}' # python3 whiteboard_to_infographic_agent.py --tool # emit the JSON tool contract _a = sys.argv[1:] if _a and _a[0] == "--tool": print(json.dumps(WhiteboardToInfographicAgent().to_tool(), indent=2)) else: _raw = _a[0] if _a else (sys.stdin.read().strip() or "{}") print(WhiteboardToInfographicAgent().perform(**json.loads(_raw))) # rci-capsule:v1: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 ```