scriptoriumadmin/fabro
1
0
Fork 0
mirror of https://github.com/fabro-sh/fabro.git synced 2026-10-12 03:51:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4
fabro/lib/components/fabro-workflow
History
Scott Werner 3c62d851fe Retry a finished run from the start 
Retry forked the source run at its last checkpoint and reran the failed
stage. It now creates a new run from the source's saved spec and starts
the workflow from the beginning in a fresh workspace, as retry did
before the Petri cutover. The new run records `retried_from` and no
`fork_source_ref`.

Retry no longer needs a checkpoint, a retained workspace, or a published
run branch, so it works for any terminal run that is not archived. To
continue from where a run stopped, fork it at a checkpoint.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-01 11:14:45 -04:00
..
src Retry a finished run from the start 2026-10-01 11:14:45 -04:00
Cargo.toml Use pebble's sandbox-driver adapter and delete fabro-pebble-sandbox 2026-09-21 17:42:15 -04:00
README.md Read workflow graphs through Petri's DOT parser 2026-09-19 12:23:57 -04:00

README.md

fabro-workflow

Fabro's platform half of a workflow run: what Fabro does around the engine.

Petri compiles and executes every run. fabro-petri is the crate that talks to the engine (fabro-dot reads a graph's shape and file references through Petri's parser), and this crate keeps what Fabro itself owns:

  • operations — creating a run around Petri's admission (materialize_admitted_run, persist_create_run), and the other run operations: fork, rewind, retry, and the timeline they resolve targets on. The run's display graph (fabro_types::RunGraph) is read off the graph Petri admitted; the DOT the workflow was written in is persisted beside it as graph_source.
  • workflow_bundle — the bundle a run is created from: every workflow of the version closure with its settings file and its files, and the RunDefinition the run records.
  • git, sandbox_git — the Git helpers a run's platform effects use, on the host and inside a sandbox.
  • pull_request — pull request creation for a finished run.
  • run_tools, services — the run tools an agent session calls.
  • web_search — the built-in web search backend.
  • run_lookup — resolving a run selector to a run.

The run records and status vocabulary are fabro_types'. Workflow diagnostics are Petri's: fabro validate, fabro preflight and the create handler report Petri's codes (attractor.*, unsupported.*, deprecated.*, info.*), plus Fabro's fabro.model.no_ready_provider when a model node has no provider ready to run it.