How to Sketch a LangChain LCEL Chain from a Tool List
LCEL sketches invent web tools. Steps only from the tool list.
The matching generator is the LangChain LCEL Chain Sketch from a Tool List (No Invented APIs) (Langchain Lcel Chain Sketch From A Tool List No Invented Apis) prompt. Browse related cards in the PromptDig library (Browse more prompts). When a filled run survives, share the version you actually use (Share a prompt).
Map each tool purpose
Sketch a LangChain LCEL chain from a pasted tool list. Never invent tool APIs, model SKUs, or benchmark scores. Start by filling Inputs, not by asking the model to remember last week's run. If a field is blank, write NONE or NOT IN INPUTS and leave it blank through Generate. The card is built so the model cannot honestly invent a number, owner, URL, or command that you did not paste.
Paste these fields before you hit run:
Tool list (name + purpose): [Tools]
Model name I lock (or UNKNOWN): [Model]
LangChain version I lock (or UNKNOWN): [Version]
Words I must not use: [Banned]
What I must never invent: [Never]
Output format: [Format]
Language: [Lang]
Max steps: [Max]
That inventory is the honesty ledger. Anything that does not appear there is forbidden in the draft. If you catch yourself adding a nice-to-have after the run, you are no longer using the card. You are ghostwriting. Put the extra fact in Inputs and run again.
Leave model UNKNOWN if blank
Generate is numbered on purpose. Do not skip a step because the first paragraph looked done. The early steps exist to stop later prose from smuggling claims.
Walk the Generate list in order:
- Honesty ledger: Tools count, Model, Version, Lang, Max. Forbidden: invented APIs, invented SKUs, invented tok/s.
- Tool map: name -> purpose from Tools.
- LCEL sketch: up to Max steps. Pseudocode pipes. Tools only from list. Model UNKNOWN if blank.
- Step list: each step input/output types at a high level without inventing schemas.
- Refuse list: invented HTTP tools, invented OpenAI SKUs beyond Model, invented eval scores.
- Diff notes: Banned/Never cuts.
- Version/model note.
- Compliance pass: Banned/Never hits. Step count vs Max. Gaps list of five. Format as Format.
If a step asks for a version lock, quote the version from Inputs in the output. If a step asks for a refuse list, keep the refuse list in the published artifact, not in a sidebar you delete. Reviewers should see what the model was not allowed to do.
Refuse invented latency numbers
Most failures are the same shape: a missing field gets a confident fill. A conversion rate appears. A Gradle task appears. A flash point appears. A caption appears on a job that asked for slide text only. Your review is to search the draft for numbers, names, and commands, then grep Inputs. No match means cut.
Honor the constraints as hard stops, not vibes:
- LCEL sketch from Tools only. Not Semantic Kernel and not AutoGen.
- Never invent APIs, SKUs, or benchmarks.
- Honor Max steps.
- No emojis.
When the card says not legal advice, not certification, not an exam dump, or not a caption engine, that sentence belongs at the top of the output. Deleting it to look more finished is how you inherit risk.
Cap the step count
Finish with the compliance pass the prompt already asks for. Quote the banned-word hits. Cut them. Print character counts when the job has a cap. Print word counts when the job has a budget. List gaps as gaps. Five missing facts are more useful than one smooth paragraph.
Tags on the card (langchain lcel sketch, no invented apis, lcel chain from tools) are a reminder of the job shape, not an invitation to wander into a neighboring cluster. If you need a different surface, open a different PromptDig card rather than stretching this one.
Fill the card, then run
Replace every bracket. Run on ChatGPT, Claude, or Gemini. Read the ledger first, then the artifact. If the model invents a commit, KPI, DOI, PEL, bid, or logo, discard the run. Tighten Inputs. Run again. Share the filled card that survived, not the first draft that sounded done.