You are helping me build the following AI agent workflow.

## Goal
Personalized Proposal Generation via LLM: Generates tailored proposals for each prospect by combining CRM data, chat transcripts, and internal notes into a specific template.

## Specification
- What it does: You get tailored proposals for each prospect that directly address their specific pain points and priorities. This happens automatically as the agent combines your CRM data, chat transcripts, and internal notes into a specific template using an LLM.
- Trigger: Run manually (Manual · on demand)
- Autonomy: You stay in control
- Expected setup effort: about an afternoon
- Tools/services involved:
  - LLM: Generate personalized proposal text based on inputs
  - CRM: Store and retrieve prospect information

## Known pitfalls, handle each one explicitly in your implementation
1. LLMs may ignore negative constraints if they are buried in long context windows so place critical exclusions at the very beginning of the system prompt.
2. Personalization tokens might contain special characters that break JSON formatting so sanitize all input strings before embedding them in the payload.

## Reference implementation
https://www.youtube.com/watch?v=5hZCTc_mwOg (workflow template)
Fetch and inspect this before building. If it matches my stack, adapt it;
if not, rebuild the pattern with my tools.

## Process requirements
1. Before building: ask me which of the listed tools I actually use and
   what my platform is (n8n / Make / code / other). Do not assume.
2. Adapt the pattern to my answers; do not force the reference stack.
3. Address every pitfall above; tell me how you handled each.
4. Provide a test plan I can run before letting this touch real data.
5. Ask before any step that sends messages, modifies data, or spends money.

Source: https://usecasesforagents.com/use-case/personalized-proposal-generation-via-llm/ via usecasesforagents.com