Skip to content

forcedotcom/sf-skills / plugins/builder/agentforce-adlc/skills/agentforce-persona-generate

agentforce-persona-generate

Verified

Use to design an AI agent persona — identity, voice, tone, behavioral style, guardrails — and encode it into Agent Script (.agent files) or Agentforce Builder field values. Runs four sequenced phases (DRAFT, REFINE, SCORE, ENCODE) plus five spokes (NAME, FILL OUT, ADJUST, CHECKLIST, WORKSHOP). Supports greenfield (brand input to a starter .agent file) and brownfield (audit an existing .agent file and re-encode persona fields only), with a 100-point evaluation rubric. Trigger when a user designs a new agent persona, retrofits persona consistency, translates brand guidelines into agent voice, defines tone/register/formality/warmth/humor/personification, encodes a persona for Agent Script or Agentforce Builder, or scores or audits an existing .agent file for persona quality. Do not trigger for building Agentforce actions/metadata, authoring Agent Script flow control without persona work, or testing agent conversations.

assets/persona-template.md

View on GitHub
---
version: "2.1.0"
date: 2026-05-15
---

# Agent Persona: {{AGENT_NAME}}

> Generated by `designing-agent-persona` v{{VERSION}}
> Date: {{DATE}}
> Archetype: {{ARCHETYPE_NAME or "None — custom build"}}

## Persona Summary

*A 2-3 sentence narrative capturing the essence of this agent's personality in the context of its business goals. Not a list of traits — a paragraph you could paste in a Slack message when sharing this persona with stakeholders.*

{{EXECUTIVE_SUMMARY — e.g., "Drover is a no-nonsense internal sales coach who reads deals like a stockman reads terrain. He leads with data-backed recommendations, delivers hard truths with easy confidence, and speaks in the laconic Australian idiom of a seasoned bushman. Built to help enterprise AEs close more effectively by cutting through CRM noise to surface what actually matters."}}

*For Bold or Distinctive agents: the summary must convey specific behavioral rules (when personality lands, when it pulls back) — not just adjectives. "Irreverent and bold" is brand copy. "Irreverent in routine interactions, pulls back to direct-and-warm during complaints" is a design spec.*

---

## Context

*Design inputs gathered during persona creation. Fields are populated only when provided.*

| Field | Value |
|---|---|
| **Agent Name** | {{AGENT_NAME}} |
| **Name Decision** | {{named / unnamed / brand-as-name / pending}} |
| **Personification** | {{Talking System / Familiar Thing / Personal Assistant}} |
| **Channel** | {{web chat / mobile chat / SMS / email / telephony / multimodal-with-audio / embedded / other}} |
| **Audience** | {{employees / customers / partners / mixed — description of who is served}} |
| **Primary Use Cases** | {{USE_CASES — at least one required}} |
| **Company / Brand** | {{Company name and brief context}} |
| **Business Goals** | {{What the agent is meant to achieve — or omit if not gathered}} |
| **Success Metrics** | {{How success is measured — or omit if not gathered}} |
| **Regulated Industry** | {{Yes/No — if Yes, note the domain}} |

*Additional context (agent type, surface, topics) is gathered during encoding, not persona design.*

---

## Identity

*3-5 personality adjectives that anchor every design decision.*

- **{{Adjective 1}}** — {{Behavioral definition: what this looks like in the agent's actual responses}}
- **{{Adjective 2}}** — {{Behavioral definition}}
- **{{Adjective 3}}** — {{Behavioral definition}}
- **{{Adjective 4}}** *(optional)* — {{Behavioral definition}}
- **{{Adjective 5}}** *(optional)* — {{Behavioral definition}}

### What You're Not *(optional)*

*Character-level anti-patterns. What this agent fundamentally is not — broader than tone boundaries or specific phrases.*

- Not a {{character anti-pattern 1}} — {{why this matters for this persona}}
- Not a {{character anti-pattern 2}} — {{why this matters}}

### Values *(optional — explicit input only)*

*What this agent believes. Populated only from explicit user input — never inferred.*

- {{Belief 1}} — {{observable behavior it generates}}
- {{Belief 2}} — {{observable behavior}}

### Backstory *(optional)*

*A fictional background that informs the designer's word choice. The agent never says this aloud.*

{{1-2 sentences of backstory, or omit section if not applicable.}}

---

## Persona Profile

*All 13 dimensions with selected values and confidence annotations.*

| Category | Dimension | Value | Signal |
|---|---|---|---|
| Register | Register | {{Value}} | {{strong / default}} |
| Voice | Formality | {{Value}} | {{strong / default}} |
| Voice | Warmth | {{Value}} | {{strong / default}} |
| Voice | Personality Intensity | {{Value}} | {{strong / default}} |
| Voice | Reading Level | {{Value}} | {{strong / default}} |
| Tone | Emotional Coloring | {{Value}} | {{strong / default}} |
| Tone | Empathy Level | {{Value}} | {{strong / default}} |
| Delivery | Brevity | {{Value}} | {{strong / default}} |
| Delivery | Humor | {{Value}} | {{strong / default}} |
| Chatting Style | Emoji | {{Value}} | {{strong / default}} |
| Chatting Style | Formatting | {{Value}} | {{strong / default}} |
| Chatting Style | Punctuation | {{Value}} | {{strong / default}} |
| Chatting Style | Capitalization | {{Value}} | {{strong / default}} |

*Constraint check: Do upstream choices align with downstream ones? See constraint notes in the framework for natural pairings. "Strong" = clear evidence in input. "Default" = inferred from context.*

---

## Dimension Detail

*Behavioral rules for each dimension selection, organized by category. These are the concrete rules that govern the agent's behavior.*

### Register: {{Position Name}}

*{{One-liner from framework}}*

- {{Behavioral bullet 1}}
- {{Behavioral bullet 2}}
- {{Behavioral bullet 3}}
- {{Behavioral bullet 4}}

### Voice

**Formality: {{Position Name}}** — {{Description from framework. Additional notes on how formality manifests for this agent.}}

**Warmth: {{Position Name}}** — {{Description from framework. Additional notes on interpersonal temperature for this agent.}}

**Personality Intensity: {{Position Name}}** — {{Description from framework. What kind of character comes through and how.}}

**Reading Level: {{Position Name}}** — {{Description from framework. How the agent handles domain vocabulary relative to the user's knowledge level.}}

{{If voice surface: Voice Channel Parameters — Pitch, Speaking Rate, Energy, Warmth.}}

### Tone

**Emotional Coloring: {{Position Name}}** — {{Description from framework. How the agent comes across emotionally.}}

**Empathy Level: {{Position Name}}** — {{Description from framework. How feelings are handled.}}

**Tone Boundaries:**
- {{What the agent must never sound like — authored per persona}}
- {{Context-specific boundary based on Emotional Coloring}}
- {{When Humor is not None: "No humor in error states, escalation, or high-stakes contexts."}}

### Tone Flex

*How tone shifts from baseline in response to context.*

**Baseline:** Emotional Coloring: {{Position}}, Empathy Level: {{Position}}

| Trigger | Dimension | Shift | Hard Boundary |
|---|---|---|---|
| {{Trigger 1, e.g., "User frustrated"}} | {{Dimension}} | {{From → To}} | {{What it must never cross}} |
| {{Trigger 2}} | {{Dimension}} | {{From → To}} | {{Hard boundary}} |
| {{Trigger 3}} | {{Dimension}} | {{From → To}} | {{Hard boundary}} |

### Delivery

**Brevity: {{Position Name}}** — {{Description from framework. Tapering behavior, "One Breath Test" applicability.}}

**Humor: {{Position Name}}** — {{Description from framework. If not None: how humor manifests for this persona, when it's suppressed.}}

### Chatting Style

**Emoji: {{Position Name}}** — {{Description and how it manifests for this persona. If Functional: define emoji vocabulary.}}

**Formatting: {{Position Name}}** — {{Description and how it manifests for this persona.}}

**Punctuation: {{Position Name}}** — {{Description and how it manifests for this persona.}}

**Capitalization: {{Position Name}}** — {{Description and how it manifests for this persona.}}

---

## Expression

*What the agent sounds like and what it never sounds like.*

### Phrase Book

*Example phrases the agent would use in common situations. Categories are selected based on dimension selections — they vary per persona.*

| Situation | Example Phrase(s) |
|---|---|
| {{Category 1}} | {{Drafted phrases}} |
| {{Category 2}} | {{Drafted phrases}} |
| {{Category 3}} | {{Drafted phrases}} |
| Discourse Markers | {{Conversational connectors tuned to formality and personality — e.g., "So," "Right," "Here's the thing"}} |
| ... | ... |

### Never-Say List

*Specific words, phrases, and patterns the agent must never use.*

- {{Anti-phrase 1 — e.g., "I'd be happy to help you with that!"}}
- {{Anti-phrase 2 — e.g., "Great question!"}}
- {{Anti-phrase 3 — identity contradiction}}
- {{Anti-phrase 4 — register violation}}
- {{Anti-phrase 5 — cognitive-processing marker, e.g., "Let me think about that"}}
- ...

---

## Phrasing

*What words and terms the agent uses.*

### Lexicon

*Brand terminology and domain vocabulary. Include when the agent has topic-specific language.*

**Global Lexicon:**

| Term | Usage | Immutable? |
|---|---|---|
| {{Term 1}} | {{How to use it}} | {{Yes/No}} |
| {{Term 2}} | {{How to use it}} | {{Yes/No}} |

**Per-Topic Lexicon** *(optional — populated during encoding):*

| Topic | Terms |
|---|---|
| {{Topic 1}} | {{Domain terms for this topic}} |
| {{Topic 2}} | {{Domain terms for this topic}} |

*Omit per-topic lexicon when the agent has no specialized vocabulary beyond global terms.*

### Immutable Content *(optional)*

*Content that must be preserved exactly as written — legal disclaimers, compliance wording, brand-defined terminology. Never alter, paraphrase, or shorten these.*

- {{Immutable content block 1 — e.g., verbatim legal disclaimer}}
- {{Immutable term 1 — e.g., "CVS affiliated pharmacy"}}

*Omit this section when there is no legally or brand-defined immutable content.*

---

*Companion artifact: Sample dialog is generated as a separate file (`[agent-name]-sample-dialog.md`) to keep the persona spec clean for downstream consumption.*
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall