global/deep-research

Deep Research — multi-source fan-out research with parallel web, YouTube, Reddit, and RSS.

Category: global
Source: workflows/research/deep_research.py

Input Schema

Field	Type	Default	Description
`depth`	`integer`	`5`
`query`	`string`	`""`
`regenerate`	`object`	—	When set, this run is a regeneration. Workflows may read `direction` / `keep` / `extra_instructions` to modulate prompts; the engine persists `parent_run_id` and `parent_variant_index` as run lineage columns.
`rss_feeds`	`object`	—
`sources`	`any[]`	—
`variants`	`integer`	`1`	Number of independent variant executions (1–10). When > 1, the engine runs the workflow N times with different sampling, producing N outputs.

Output Schema

No schema defined.

Task Pipeline

plan_research → search_web → read_webpages → research_youtube → fetch_rss_feeds → search_reddit → merge_research → synthesize_research → format_report

Task	Description
`plan_research`	Validate input and decide which sources to activate.
`search_web`	Search the web via Exa (semantic search) or Jina Search (free fallback).
`read_webpages`	Read top web results as clean markdown via Jina Reader (free, no API key).
`research_youtube`	Search YouTube for relevant videos and extract metadata + transcript excerpts.
`fetch_rss_feeds`	Parse RSS/Atom feeds for latest content.
`search_reddit`	Search Reddit via shared Reddit client (old.reddit.com, UA rotation, retry).
`merge_research`	Join function — merge parallel branch outputs into a unified context.
`synthesize_research`	Analyze all collected sources with Gemini and produce a structured synthesis.
`format_report`	Structure the final output, stripping internal keys.

Run-spec example

Save the YAML below as my-run.yaml, edit the values, and run with the CLI or POST it to the API. Required fields are uncommented; optional knobs are documented above the input: block — copy any line under input: and uncomment to set.

workflow: global/deep-research

# Optional fields — copy any line(s) under `input:` and uncomment to set:
#   depth: 5
#
#   query: ""
#
#   rss_feeds: null
#
#   sources: []
#

input: {}

Run it locally:

fab-workflow --from-file my-run.yaml

Or submit over the wire — the same file is the request body:

curl -X POST 'https://gofabric.dev/v1/workflows/runs?name=global/deep-research' \
  -H 'Authorization: Bearer fab_xxx' \
  -H 'content-type: application/yaml' \
  --data-binary @my-run.yaml

Every workflow also accepts the universal WorkflowInput fields — variants (1–10 fan-out) and regenerate (creative-direction hints with run lineage). See Run-specs (YAML / TOML / JSON) for the full top-level shape (metadata, priority, bundle, parent, etc.).

Warnings

Last user task format_report has no Pydantic return type — workflow output schema is null. Declare a WorkflowOutput subclass and pass it to Flow(output=…) for a strict contract.
Task merge_research has no Pydantic types — contract is opaque to consumers.