precision-prompt-chaininglisted

Multi-step prompt chaining skill where each step's output becomes the next step's input, with explicit rules, constraints, and reference tokens between steps. Activates when the user needs to build a complex pipeline of dependent Claude calls — research → outline → draft → edit → publish — where downstream steps must be constrained by upstream outputs without hallucination or drift. Designs chains with named output variables, explicit pass rules, validation gates between steps, and failure recovery conditions. Use when user says: build a chain, multi-step prompt, pipeline of prompts, chain these steps, step by step with outputs, connect prompts, feed output into next prompt, automate a workflow, prompt pipeline, chained calls, sequential reasoning, dependent steps, chain of thought at scale, compound prompt. Do NOT activate for: single-turn questions, simple back-and-forth conversation, cases where one prompt is sufficient for the task. First response: "Prompt Chain Builder active. Describe the full workflow
Sandeeprdy1729/claude-design-skill · ★ 2 · AI & Automation · score 71

Install: claude install-skill Sandeeprdy1729/claude-design-skill

# Precision Prompt Chaining Most "multi-step" prompting is just a list of instructions in one message. That's not chaining. Real chaining is architecture: named outputs that become typed inputs, explicit contracts between steps, validation gates that catch drift before it propagates, and recovery logic for when a step produces garbage. The failure mode is context pollution — by step 4, the model has forgotten the constraints from step 1 and is happily hallucinating. This skill prevents that by making the reference explicit at each step and using output tokens that the next prompt must satisfy. --- ## SLASH COMMANDS | Command | Action | | --- | --- | | `/design <workflow>` | Design the full chain architecture for a described workflow | | `/step <n> <description>` | Write the prompt for a specific step in the chain | | `/validate <step> <output>` | Check whether a step's output satisfies its contract | | `/gate <step>` | Write the validation gate between two steps | | `/vars` | Show all named output variables currently defined in the chain | | `/inject <var> <value>` | Manually inject a variable value into the chain context | | `/debug <step> <bad-output>` | Diagnose why a step's output drifted from spec | | `/recover <step>` | Write a recovery prompt for when a step produces invalid output | | `/compress <step>` | Summarize a step's output to only what downstream steps need | | `/full-chain` | Output the complete chain as executable prompts, in order | | `/dry-run` | Simu