AI Toolbox
Free Prompt Optimizer for Claude (Fable 5)
Paste a prompt, choose where you will run it, and get deterministic suggestions grounded in official Claude prompting guidance. Your prompt text never leaves this browser.
Lint runs locally; we do not see your prompt text. These are deterministic heuristics, not a verdict. Use them as a checklist before shipping.
Lint runs locally; we do not see your prompt text.
These are deterministic heuristics, not a verdict. Use them as a checklist before shipping. Future rewrite modes will be opt-in and clearly separate from this local checker.
Approximate token range: 0–0
Run the linter to see severity-tiered findings.
Official snippets library
Copy small, placement-labelled snippets into your system prompt, user message, or harness.
System prompt
Anti-overplanning
Discourage unnecessary planning loops.
Prefer the smallest plan that lets you act safely; update it only when new facts change the path.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
System prompt
Anti-overengineering
Keep implementations simple.
Choose the simplest implementation that satisfies the stated constraints and tests.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
User message
Brevity
Request concise output.
Answer in the fewest words that preserve the key decision, evidence, and next step.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
Harness
Checkpoint
Force progress checks in agent loops.
After each major step, summarize what changed, what was verified, and the next action.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
Harness
Grounded progress
Report only verified progress.
Do not claim success until a command, test, or observable artifact verifies it.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
System prompt
Autonomy
Permit safe independent action.
When the next step is reversible and within scope, take it without asking.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
System prompt
Boundaries
Clarify what not to modify.
Stay within the requested scope; if required information is missing, state the blocker instead of guessing.- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
Rules this linter checks
Issues
multishot-examples
Add multishot examples
Examples are the most reliable way to steer exact output shape.
Format-sensitive task without examples. Add 3–5 diverse examples in <example> tags.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
long-context-ordering
Move long context before instructions
Long-context prompts are easier to ground when evidence is loaded before the ask.
Large pasted context appears after the task. Put documents first, then the query at the end.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
reasoning-echo
Avoid chain-of-thought echo requests
Echoing hidden reasoning can trigger reasoning-extraction safeguards.
Prompt asks to reveal/transcribe hidden reasoning. Ask for a concise rationale or summarized thinking instead.
Citations
- Prompting Claude Fable 5 — Fable 5 responds better to concise intent and appropriate effort guidance than behavior over-prescription.
Suggestions
structure-sections
Add explicit sections
Structure helps the model separate source material, task, and expected output.
Long prompt with no section markers. Use <context>, <task>, and <output_format> or clear headings.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
negative-instructions
Rewrite negative instructions
Positive behavioral targets are easier to follow than dense prohibitions.
High density of do-not/never instructions. Say what to do instead of only what to avoid.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
constraint-without-rationale
Explain hard constraints
Rationale improves compliance with unusual constraints.
ALWAYS/NEVER/MUST without nearby rationale. Add because/so that language explaining why the rule matters.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
missing-output-format
Specify output format
Explicit output contracts reduce unusable responses.
Prompt does not state the desired output format. Name the format: JSON, table, markdown, bullets, or exact fields.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
vague-qualifiers
Replace vague quality words
Measurable targets make subjective editing requests actionable.
Vague words such as better/professional without criteria. Use measurable criteria, rubrics, examples, or constraints.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
missing-context-intent
Add context and intent
Audience and purpose help choose tone, depth, and trade-offs.
Task-only prompt lacks audience, goal, or purpose. Use “I am working on X for Y because Z.”
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.
caps-emphasis
Normalize all-caps emphasis
Aggressive emphasis can overtrigger modern Claude models.
Excessive CRITICAL/MUST/NEVER style emphasis. Use calm wording and one explicit priority statement.
Citations
- Prompting Claude Fable 5 — Fable 5 responds better to concise intent and appropriate effort guidance than behavior over-prescription.
Info
fable-overprescription
Simplify Fable behavior steering
Fable is more reliable with concise intent than enumerated behavior laws.
Repeated always/never/must behavior steering. Prefer a short instruction describing the desired behavior.
Citations
- Prompting Claude Fable 5 — Fable 5 responds better to concise intent and appropriate effort guidance than behavior over-prescription.
fable-effort-guidance
Set effort on supported surfaces
Effort is model/surface-specific and irrelevant in claude.ai.
Complex work on API/Claude Code can benefit from effort guidance. Use high by default; consider max for very complex agentic work. Never set effort for Haiku.
Citations
- Prompting Claude Fable 5 — Fable 5 responds better to concise intent and appropriate effort guidance than behavior over-prescription.
prompt-bloat
Trim prompt bloat
The concept asks for local token-range feedback; bloat detection turns that estimate into an actionable rule.
Very long prompt appears repetitive or boilerplate-heavy. Keep durable constraints, remove repeated boilerplate, and move reusable policy to system or harness context.
Citations
- Anthropic prompt engineering guide — Claude performs best when prompts are explicit, structured, and grounded with examples where needed.