---
name: cascading-research-pipeline
description: "Orchestrates FTR-28 across a dependency graph of research steps — parent chain ID with numbered sub-steps, cascading and parallel dependencies, a pause-and-confirm gate so a step's output is actually loaded into context before anything depending on it runs, and a full worked example. Use when a person asks to install or apply \"Cascading Research Pipeline\" (FTR-30) in a project, or when working in a project that has it installed."
---

# Cascading Research Pipeline (FTR-30, version 1.1)

> Generated from library record FTR-30 version 1.1, publish 20261102a. Review status: unreviewed. The complete specification is in references/spec.md and is the authority; this file is the operating procedure drawn from it.

## What this does

Orchestrates FTR-28 across a dependency graph of research steps — parent chain ID with numbered sub-steps, cascading and parallel dependencies, a pause-and-confirm gate so a step's output is actually loaded into context before anything depending on it runs, and a full worked example

## When to use this skill

- A person asks to install, apply or update "Cascading Research Pipeline" (FTR-30) in a project.
- A project's agent instructions list cascading-research-pipeline as installed and the task touches what it covers.
- Requires first: FTR-28 (skill multi-ai-research-synthesis).
- Works better with, if installed: FTR-29 (skill legal-case-research-synthesis).

## Install procedure

Steps run in order. Assess comes first, every question in the Ask step is answered by the person before anything is written, and every write merges additively into the existing project files (read the file first, update an existing section in place, never overwrite).

### Assess first

- Confirm FTR-28 is installed — this feature has no independent engine; every step in a chain is a normal FTR-28 research item underneath.
- If FTR-29 is also installed and the chain's subject matter is legal casework: confirm whether this chain belongs on `LR` or `RS` per FTR-29's track-assessment guidance, at the chain level.
- Check the Research Register (or Master Case File) for any chain already in progress before starting a new one with a similar subject — avoid creating a duplicate chain for research that's already partway done elsewhere.
- Before executing any queued step: confirm its dependencies (if any) are actually CLOSED, not just FILED — do not proceed on a dependency still sitting at PENDING CONTEXT LOAD.
- Before executing a triangulation-preferred step specifically: confirm deep research capability is active (see the guardrail above).

### Ask, only if needed

- When a chain is proposed from a goal description: confirm the proposed step list, dependency graph, and parallel groups before assigning any IDs or writing any prompts.
- When an explicit list of prompts is given with no stated relationships: ask whether any of them actually depend on each other's output, or whether they're independent items just being tracked together for convenience.
- When a step reaches FILED — PENDING CONTEXT LOAD and something depends on it: nothing to ask — state plainly what the person needs to do (load it into Project Files) and wait for their confirmation before proceeding.
- If a chain looks like it might need more than roughly eight or nine steps: confirm whether it should really be one chain, or whether part of it is better split into its own separate chain.

### Write

The full Write payload is in references/write-payload.md. It is applied exactly as written, with the merge rule above.

### Confirm

Confirm: the chain's parent ID is assigned, every step has its sub-ID and is registered with its dependencies and parallel groups correctly reflected, the shared context primer exists, and the first runnable step (or steps, if the chain starts with a parallel group) is ready to execute. State plainly which step(s) can run first and which are waiting on a dependency.

## Other sections in the specification

- Why this exists
- Chain ID scheme
- Three ways a chain starts
- Chain generation — the decomposition step
- The dependency graph, concretely
- The state machine
- Worked example — a chain in action
- Deep-research-mode guardrail
- The filing loop
- No standalone runbook file
- Not in scope
- Interaction with FTR-29 (Legal Case Research Synthesis)

These are in references/spec.md.

## Files in this skill

- references/spec.md: the complete specification, unabridged.
- references/write-payload.md: the Write step on its own.