More screenshots
Works with the AI Agents you already use
Prompt Preflight — tokens, window and cost before you ship
- Turkish, other non-English text, code and CJK usually produce more tokens per character: treat the estimate as a lower bound.
Free
About this skill
Answer before the prompt is deployed: how many tokens, how much context window, how much money, and where does caching help?
Never ask the user to paste a key into chat. Otherwise use the heuristic and label it an estimate:
-
tokens ≈ characters / 3.7for English prose (older tokenizer). -
Models on the newer tokenizer (5.x family, Opus 4.7/4.8) count the same text about 30% higher: multiply by 1.3.
-
Turkish, other non-English text, code and CJK usually produce more tokens per character: treat the estimate as a lower bound.
-
Cached input (cache read) costs
cache_read_rateinstead ofinput_rate. Show both cold and steady-state cached cost. -
The first call that writes the cache costs 1.25x the input rate (5-minute TTL).
-
Haiku 5.5 has two rate cards: a higher one above 100K prompt tokens.
Without an API key, token counts are heuristic estimates (lower bound for non-English text and code). The model price table is a dated snapshot (October 2026); verify against Anthropic's pricing page.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 3 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts
Only contacts api.anthropic.com (free count_tokens endpoint) if the user agrees and already has ANTHROPIC_API_KEY in their environment. Otherwise fully offline heuristics.