CHUNKMAPS

ChunkMap Persuasion Flow

Structure pitches, leadership asks, and presentations with Chunk Maps, Journey paths, counterarguments, and CHUNKS_OK.

Skills library · Persuasion

Audience: leaders, sellers, educators, creators, coaches, and agents drafting persuasive communications for human-to-human influence.

What this skill does

ChunkMap Persuasion Flow guides an agent (or human) to structure logical argument chains, map counterarguments, and design persuasive narrative arcs for sales, leadership, mediation, or presentations using v1 ChunkMaps.

When to use: pitches, team alignment, essays, speeches, whiteboard sessions, and conflict mediation-anywhere agreement or action is the goal.

Do not use this skill alone for multi-agent orchestration or policy compilation. See the agentic skills for those jobs.

InputRequiredNotes
AudienceYesWho must agree / act
DomainYesArea of interest (keeps chunking relevant)
Desired outcomeYesEnd-point of the journey (what agreement looks like)
Current situationYesAudience starting state
ConstraintsNoTime, taboo topics, brand voice, legal claims
Known objectionsNoCounterarguments to map in advance
Evidence / proof pointsRecommendedFacts, stories, credentials

Step-through: the .md procedure

  1. Set Domain and Desired Outcome.
    • Domain = topic boundary (e.g. Team meeting culture).
    • Desired Outcome = the agreement or action you want (e.g. Adopt a 25-minute decision-memo format).
  2. Build a Chunk Map of the Domain with three moves:
    1. Chunk up - higher meaning, purpose, shared values, policy intent.
    2. Chunk down - specifics, evidence, examples, procedures.
    3. Chunk across - peer alternatives, related cases, lateral options.

    Keep top-level siblings within working-memory limits (7±2). Stay inside the Domain.

    DOMAIN: <topic>
    DESIRED OUTCOME: <agreement / action>
    ├── [UP] abstract / purpose chunks
    ├── [ACROSS] sibling options / related cases
    └── [DOWN] evidence / examples / details
  3. Plot a Journey of Persuasion. Do not dump the map linearly. Design a path that:
    1. Starts near the audience current knowledge (reduces confusion).
    2. Chunks up to shared purpose / risk / value when needed.
    3. Chunks down to concrete proof after the frame is accepted.
    4. Chunks across to address alternatives or objections.
    5. Ends at the Desired Outcome with clear next steps.

    Link new ideas to existing knowledge (leverage natural TDS) via links of agreement, not leaps.

  4. Map counterarguments. For each likely objection, name where it lives on the map, the response move (up / down / across), and the supporting chunk.
    ObjectionWhere on mapResponse moveSupporting chunk
    Too expensiveAcross: cost alternativesChunk up to cost of inaction; down to ROI proofPilot cost vs lost decision time
    We have always done it this wayDown: current habitsChunk up to shared goal; across to peer successPeer team shipping decisions in half the meetings
  5. Encode the narrative with CHUNKS_OK (one or two sentences per letter):
    LetterPrompt
    CCurrent Situation your audience is in
    HHeadaches caused by the current situation
    UUtopia wanted by your audience for the future
    NNext Steps - what they should do immediately
    KKnowledge - why you can help / why your solution / why they benefit
    SSolution Overview - how you solve the headaches
    OOffer - compelling reason to accept now
    KKnowledge (close) - reinforce credibility and benefit

    Order of delivery may vary by channel; the letters ensure completeness.

  6. Deliver and adapt. Keep the rich Chunk Map available. If the audience stalls, take an alternate route through the map (up / down / across) without abandoning the Desired Outcome.

Output schema

The skill expects a persuasion pack shaped like:

# Persuasion pack: <title>

## Meta
- Audience / Domain / Desired outcome / Channel

## Chunk Map (outline)
## Journey path (ordered stops)
## Counterargument map
## CHUNKS_OK draft
## Open risks

Full example prompt - copy and use

Note: You must edit the User Inputs section before running the example prompt. You may paste the EXAMPLES below into the prompt USER INPUT area, or replace them with your own audience, domain, and objections.

USER INPUT (replace placeholders)

Use these slots in the paste-ready prompt. The EXAMPLES are a ready-made leadership persuasion trial: convince department heads to adopt a 25-minute decision-memo meeting format.

## USER INPUT (replace placeholders)
<<PASTE audience>>
<<PASTE domain>>
<<PASTE desired_outcome>>
<<PASTE current_situation>>
<<OPTIONAL: constraints>>
<<OPTIONAL: known_objections>>
<<RECOMMENDED: evidence_proof_points>>

EXAMPLES:
audience:
---
"Department heads in a 120-person product company who own weekly status meetings"
---

domain:
---
"Team meeting culture"
---

desired_outcome:
---
"Approve a 90-day pilot of 25-minute decision-memo meetings (one memo in advance; meeting ends with a written decision)"
---

current_situation:
---
"Weekly status meetings run 60-90 minutes, decisions slip to email, and the same topics recur without closure."
---

constraints:
---
- "No blame language about individuals"
- "Stay inside meeting process; do not pitch unrelated tools"
- "Pilot must fit existing calendar without adding a new standing meeting"
---

known_objections:
---
- "We do not have time to write memos"
- "Async will kill healthy debate"
- "This is just another process fad"
---

evidence_proof_points:
---
- {id: "EV-01", claim: "Last quarter, 14 of 18 'decided' items were reopened in a later meeting", source: "PMO tracker export"}
- {id: "EV-02", claim: "Average status meeting length is 78 minutes across 6 teams", source: "Calendar sample, 6 weeks"}
- {id: "EV-03", claim: "Peer squad cut meeting hours 35% after decision memos", source: "Internal Ops retrospective Q2"}
---

The Persuasion Flow Prompt

Paste the controller prompt below into an LLM (system message, or first user message). The USER INPUT section inside the prompt already includes the EXAMPLES so you can run a trial immediately - edit them first if you have a real persuasion brief. The agent must produce a Chunk Map, Journey path, counterargument map, and CHUNKS_OK draft - not one flat pitch essay.

Get Raw Prompt
SYSTEM / CONTROLLER PROMPT - ChunkMap Persuasion Flow

You are a ChunkMaps persuasion agent. Do NOT produce a single unstructured pitch essay.
You must follow the ChunkMap Persuasion Flow procedure: gather inputs, build a Domain Chunk Map,
plot a Journey of Persuasion, map counterarguments, encode CHUNKS_OK, and list open risks.

## Mission
Serve Desired Outcome from USER INPUT. Default trial:
Domain = Team meeting culture
Desired Outcome = Approve a 90-day pilot of 25-minute decision-memo meetings
Audience = Department heads who own weekly status meetings

## Hard rules
1. Stay inside Domain. Prune off-topic chunks.
2. Keep top-level siblings within 7+/-2.
3. Never invent evidence. Use only evidence_proof_points / USER INPUT; otherwise flag open risks.
4. Link new ideas with links of agreement (TDS), not leaps.
5. If audience stalls mid-path, propose an alternate up/down/across route without changing Desired Outcome.
6. Offer (O) only after Headaches (H) are clear.
7. Final deliverable MUST use the Output schema below.

## Procedure (run in order)
### Step 1 - Confirm inputs
Restate Audience, Domain, Desired Outcome, Current Situation, Constraints, Known objections, Evidence.

### Step 2 - Build Chunk Map
Use UP / DOWN / ACROSS. Emit outline form:
DOMAIN: ...
DESIRED OUTCOME: ...
├── [UP] ...
├── [ACROSS] ...
└── [DOWN] ...

### Step 3 - Journey of Persuasion
Ordered stops from audience-familiar start chunk to Desired Outcome.
Label each stop with move: start | up | down | across | land.

### Step 4 - Counterargument map
Table: Objection | Where on map | Response move | Supporting chunk
Cover every known_objection; add likely extras only if marked as inferred.

### Step 5 - CHUNKS_OK draft
Fill C, H, U, N, K, S, O, K(close) with 1-2 sentences each, grounded in the map and journey.

### Step 6 - Open risks
List claims needing evidence and tone risks.

## Output schema (emit exactly these sections)
# Persuasion pack: <title>

## Meta
- Audience:
- Domain:
- Desired outcome:
- Channel: (deck / email / meeting / speech)

## Chunk Map (outline)
...

## Journey path (ordered stops)
1. Start chunk -> ...
n. Desired outcome

## Counterargument map
| Objection | Move | Response |

## CHUNKS_OK draft
- C:
- H:
- U:
- N:
- K:
- S:
- O:
- K (close):

## Open risks
- Claims needing evidence:
- Tone risks:

## USER INPUT (replace placeholders)
# Edit before running. EXAMPLES included for a trial run.
audience:
---
"Department heads in a 120-person product company who own weekly status meetings"
---

domain:
---
"Team meeting culture"
---

desired_outcome:
---
"Approve a 90-day pilot of 25-minute decision-memo meetings (one memo in advance; meeting ends with a written decision)"
---

current_situation:
---
"Weekly status meetings run 60-90 minutes, decisions slip to email, and the same topics recur without closure."
---

constraints:
---
- "No blame language about individuals"
- "Stay inside meeting process; do not pitch unrelated tools"
- "Pilot must fit existing calendar without adding a new standing meeting"
---

known_objections:
---
- "We do not have time to write memos"
- "Async will kill healthy debate"
- "This is just another process fad"
---

evidence_proof_points:
---
- {id: "EV-01", claim: "Last quarter, 14 of 18 'decided' items were reopened in a later meeting", source: "PMO tracker export"}
- {id: "EV-02", claim: "Average status meeting length is 78 minutes across 6 teams", source: "Calendar sample, 6 weeks"}
- {id: "EV-03", claim: "Peer squad cut meeting hours 35% after decision memos", source: "Internal Ops retrospective Q2"}
---

channel:
---
"leadership meeting"
---

auto_run: true

## Start
Confirm inputs, then run Steps 2-6. End with the full Persuasion pack. Do not skip the counterargument map or CHUNKS_OK.

Next: The Journey Of Persuasion