# rfp agent hit token limit mid cover letter draft and failed to continue: how to fix it

Drafting a long cover letter in one call blows the output token budget. Draft a tight outline first, then expand one paragraph at a time within the limit.

```text
rfp agent hit token limit mid cover letter draft and failed to continue
```

## When to use this skill
- Agent stops mid cover letter with a token error
- Cover letter draft truncates

## When not to use this skill
- Draft finishes but is low quality (a prompt issue)
- Agent runs out of context, not output tokens (an input issue)

## Tool compatibility
Agent long-form drafting within token limits. Related area: model APIs.

Related area: proposal automation and agent-assisted RFP workflows.

Cloud UIs change labels over time; if a button moved, search settings for the closest match.

## Before you start
- Sign in to the tool and keep one tab open so sessions stay clean
- Copy the exact error text and note the time it happened
- Set aside 15 to 20 minutes; most of these resolve in a single pass

## Quick diagnosis
Make sure this is the right skill before you invest the 15 minutes.
- If you are dealing with agent stops mid cover letter with a token error, start at step 1 below.
- If instead draft finishes but is low quality (a prompt issue), stop here; you need a different skill.
- If neither quite matches, compare your screen against the verbatim error block at the top; that block is exactly what this skill covers.

## Fix it in steps
### 1. Draft a paragraph outline

```
Generate a short outline: one line per paragraph.
```
Expected: Expected: outline fits easily in the token budget.

### 2. Expand one paragraph at a time

```
Generate each paragraph as its own call from the outline.
```
Expected: Expected: each call completes inside the limit.

### 3. Assemble the letter

```
Join the paragraphs in outline order.
```
Expected: Expected: full letter assembled.

### 4. Polish in a final pass

```
Run one editing pass over the assembled letter.
```
Expected: Expected: consistent tone throughout.

## Confirm each step worked
Run through this checklist before moving on. If a check fails, redo that step instead of pushing ahead.
- Step 1 (draft a paragraph outline): Expected: expected: outline fits easily in the token budget.
- Step 2 (expand one paragraph at a time): Expected: expected: each call completes inside the limit.
- Step 3 (assemble the letter): Expected: expected: full letter assembled.
- Step 4 (polish in a final pass): Expected: expected: consistent tone throughout.

## Quick recap
1. Draft a paragraph outline
2. Expand one paragraph at a time
3. Assemble the letter
4. Polish in a final pass

## Still stuck?
Gather three things: the exact error text, the time it happened, and what changed just before it broke.
Check the vendor status page for an active incident. Then open a support ticket with those three things;
it cuts the back-and-forth in half. For agent runs, attach the last 50 lines of the agent log too,
since the loop or timeout signature in the log usually names the cause.

## Variant phrasings
### agent token limit cover letter
Outline-first drafting fixes it. If this matches what you saw, the steps above apply as written.

### rfp agent output tokens exhausted
Paragraph-by-paragraph expansion works. If this matches what you saw, the steps above apply as written.

## Why it happens
Output token limits cap a single generation, so the letter must be built in pieces. An outline keeps the pieces coherent; without it, per-paragraph generation drifts in tone and structure.

## Edge cases
- Set the outline first and do not let the agent skip it
- Keep each paragraph call focused; extra instructions eat the budget
- A final polish pass is cheaper than regenerating the whole letter

## Provenance

Resolved from the public thread: https://vectle.com/posts/pst_Uk28M8iCKAHXYyGv25AL3w