global/content-modify

Content modification workflow — rewrite/redesign any content type via a user prompt.

Category: global
Source: workflows/content/modify.py

Input Schema

Field	Type	Default	Description
content	string	required	The content to modify. For text, pass the full text inline. For media (image/audio/video), pass a file path or URL.
content_type	string	""	Content type override: ‘text’, ‘image’, ‘audio’, or ‘video’. Auto-detected from content if omitted.
preserve	string[]	—	Aspects of the original content to preserve unchanged. Examples: ‘tone’, ‘length’, ‘structure’, ‘style’, ‘key_points’.
prompt	string	required	Natural-language instruction describing the desired modification.
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.
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

Field	Type	Default	Description
content_type	string	required	Detected or specified content type.
fidelity_score	number	0.0	How well the modification matches the prompt (0-10).
kind	object	—	Variant card shape: video / carousel / image / text. Surfaced on the per-variant entry of the run-output API and used by gallery UIs to pick the right layout.
modifications_applied	string[]	—	List of specific changes that were made.
modified_content	string	required	The modified content. For text, the full rewritten text. For media, the file path to the modified asset.
original_summary	string	""	Brief summary of the original content.
workflow	string	""	Workflow name that produced this output.

Task Pipeline

analyze_content → apply_modifications → evaluate_and_refine

Task	Description
analyze_content	Analyze the input content to understand its structure and type.
apply_modifications	Apply the requested modifications to the content.
evaluate_and_refine	Evaluate whether the modifications match the prompt and refine if needed.

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/content-modify

# Optional fields — copy any line(s) under `input:` and uncomment to set:
# Content type override: 'text', 'image', 'audio', or 'video'. Auto-detected from content if omitted.
#   content_type: ""
#
# Aspects of the original content to preserve unchanged. Examples: 'tone', 'length', 'structure', 'style', 'key_points'.
#   preserve: []
#

input:
  # The content to modify. For text, pass the full text inline. For media (image/audio/video), pass a file path or URL.
  content: ""

  # Natural-language instruction describing the desired modification.
  prompt: ""

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/content-modify' \
  -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.).