Skip to main content
Agent skill

Prysai Dialogue Brief

Turn an untried, low-risk request for LLM help into one short, copy-ready first-turn dialogue brief with a stated outcome, audience, supplied inputs, constraints, output shape, acceptance check, and stop boundary. Use when the user has a roughly defined communication goal but needs a clear first message; do not use for coaching, research, a task with files or external effects, or repairing an already failed reply.

Get started

Get the complete skill folder, including its scripts and reference files.

When to use this skill

Use Prysai Dialogue Brief when an AI agent needs a reusable SKILL.md workflow for this job: Turn an untried, low-risk request for LLM help into one short, copy-ready first-turn dialogue brief with a stated outcome, audience, supplied inputs, constraints, output shape, acceptance check, and stop boundary. Use when the user has a roughly defined communication goal but needs a clear first message; do not use for coaching, research, a task with files or external effects, or repairing an already failed reply.

When not to use it

Skip Prysai Dialogue Brief when the task is outside the analysis category, or when a more specific skill in this directory already covers the same workflow with clearer triggers.

How to install

Where will you use it?

Install scope
Claude Code folder setup
mkdir -p ~/.claude/skills/prysai-dialogue-brief
# Save SKILL.md to ~/.claude/skills/prysai-dialogue-brief/SKILL.md
  1. Claude Code personal install: place the skill and its bundled files at ~/.claude/skills/prysai-dialogue-brief/SKILL.md. For a shared project, commit the folder at .claude/skills/prysai-dialogue-brief/ instead.
  2. Current Claude Code detects SKILL.md changes in watched directories during the session. If you just created a top-level skills directory, run /reload-skills. Check that the skill is listed before testing its trigger.
  3. For Claude chat or Cowork in the Desktop app, manage uploaded skills in Customize > Skills; copying files into a local Claude Code folder is not a Desktop install. Other agents may use different install locations.

Full install guide for Claude, Cursor, and Codex →

Try a starter prompt

After installing the skill, paste this starter prompt into your agent to verify it triggers on a small task.

Starter prompt
Use the Prysai Dialogue Brief skill for this task: Use Prysai Dialogue Brief when this documented workflow matches the task. Confirm any required inputs or files before making changes.

What supports this recommendation?

69/100 provisional content score. This measures documentation, practical detail and source signals. It is not a measured task success rate or a security certification.

Documentation
77/100
Practical detail
68/100
Source evidence
54/100
Source signals
72/100
Model compatibility evidence
  • claudeinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • chatgptinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • geminiinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • copilotinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • llamainferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • perplexityinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • mistralinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

  • grokinferred

    The skill uses model-agnostic prompt or LLM terminology. Portability is inferred and has not been verified on this model.

Inferred compatibility is based on content signals. Check the app, its available tools and setup requirements before running a skill.

What this skill does

# Dialogue Brief

Create one small first message that the user can inspect and send. This Skill owns the seam before a substantive model answer, tool action, research step, or teaching loop begins. It does not execute the request or judge the answer.

## Route before drafting

Use this Skill only if all of these are true:

- the user has not yet sent the request or received a reply to repair; - the desired result is a text-only, low-risk first conversation; - no files, tools, account access, browsing, private records, publication, or external action are needed; and - the user wants help expressing one bounded request, not help practising a skill or researching a factual question.

Yield instead of stretching the brief:

- a learner wants a baseline, feedback, correction, or transfer practice: `prysai-learning-coach`; - the request involves Codex concepts, tools, Skills, or Agent behavior: `prysai-codex-coach`; - files, tools, accounts, permissions, external actions, or a real delivery target are involved: `prysai-task-protocol`; - the user needs current facts, sources, or a source-backed conclusion: `prysai-source-investigator` or `prysai-research-router`; - the original request and an unsatisfactory reply already exist: `prysai-communication-failure-triage`; - the user asks whether an existing claim is supported: `prysai-evidence-review`.

Stop rather than requesting secrets, sensitive personal data, unpublished records, credentials, account state, or private prompts. A request to draft a brief does not authorize a later action.

## Collect only the first-turn essentials

Collect these fields, using the user's words where possible:

```text outcome: one observable result from the first reply audience: who will use or read the result supplied_inputs: the safe text or facts available for this turn constraints: facts to preserve, limits, tone, exclusions, or help policy output_shape: the requested form and length acceptance_check: what the user will inspect before accepting it stop_boundary: what must not happen or what missing fact ends the turn ```

If one missing field would materially change the result, return the `needs_clarification` receipt below with **one** plain clarifying question. Do not draft a partial brief, collect a biography, invent an audience, fill an unknown with a plausible detail, or ask several questions merely to make the brief look thorough. If the observable outcome cannot be named after that one clarification, return `blocked: outcome_not_observable` and state the smallest missing decision.

## Draft the first-turn brief

Return a 120–180-word brief followed by a copy-ready first message. Keep the scope to one turn. Use direct, ordinary language; do not add roles, emotional pressure, hidden-reasoning requests, performance promises, or generic “be helpful” filler.

The copy-ready message must contain these labeled elements in natural prose:

```text Outcome Audience Supplied inputs Constraints Output shape Acceptance check Stop boundary ```

If the user has not supplied a fact needed for the answer, ask the receiving model to label that fact `unknown` rather than inventing it. If sources are needed, request a source plan or stop; do not request a confident factual answer without evidence.

## Return a compact receipt

If a material field is missing, return exactly:

```text brief_status: needs_clarification clarifying_question: known_inputs: risk: R0 content_status: candidate handoff: ```

Once the fields are sufficient, return exactly:

```text brief_status: ready_to_copy | blocked dialogue_brief: 120–180 words first_turn: copy-ready text inputs_preserved: unknowns: acceptance_check: stop_boundary_or_blocker: risk: R0 evidence: selected brief revision only content_status: candidate handoff: ```

Accept the output only when it preserves the supplied facts, contains one observable acceptance check, forbids unstated action or data expansion, and names a route for work outside the first turn. `ready_to_copy` means only that the brief is present; it does not establish model behavior, answer quality, learning, factual correctness, user satisfaction, or completion.

## Maintenance record

- `source`: original Prysai Lab method derived from the communication-clinic, task, evidence, and routing contracts - `license`: original rewrite; external material remains reference-only under `docs/sources/asset-register.md` - `owner`: communication-systems maintainer - `version`: `0.1.0` - `review_date`: `2026-09-12` - `content_status`: `candidate`

Intended uses

  • Use Prysai Dialogue Brief when this documented workflow matches the task.

Implementation notes

  • Review the source instructions and adapt inputs before running the workflow.

Try it, then tell us how it went

A copied prompt does not tell us whether the task worked. Report your result after checking the output. We collect your tool and outcome, never your files or prompt.