Repository navigation
Plan: the job index — a job you can name, find and read back - #125
Merged
Merged
Conversation
…Job node for search, no artifact node Findings from an exploration of how jobs and artifacts are stored today (no StrutJob or StrutArtifact in the graph, `job` on run.start and the summary only, the record with no title or run list, artifacts a read-time resolution of any run's output, the title hive sends stripped by the job workflow's input block, no jobs tool on the seed), and the proposed shape: title on the launch body, run lines on the record written by the runner through the jobs capability, GET /jobs + GET /jobs/:id with a read-time artifact fold, job on a session's turn line, job/list|get|read steps for the agent, and in the graph a `job` attribute plus one StrutJob node with an IN_JOB edge. Seven choices are listed as proposed rulings, none made.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
A plan only: `plans/job-index.md`, plus one pointer line in the status header of `plans/jobs.md`. No code.
What it answers. How jobs and artifacts are stored today, and the shape for a job that can search past jobs, learn what they produced, look at their artifacts, and find the thread to continue.
Findings. There is no `StrutJob` or `StrutArtifact` node type, and `job` is an attribute on no graph node. A job is a directory plus a record with no title and no list of its runs; its runs are findable only by scanning every workflow. `artifacts[]` is resolved off any run's output at read time and stored nowhere. The title hive sends rides on `input`, which the `job` workflow's input block strips before the run records it. The seed grants no jobs or sessions tool.
The proposed shape. Title on the launch body beside `job`; one run line per launch on the record, written by the runner through the existing jobs capability; `GET /jobs` and `GET /jobs/:id` with the job's current deliverables folded newest-first by artifact id; `job` on a session's turn line; `job/list`, `job/get`, `job/read` for the agent; in the graph a `job` attribute on `StrutRun` and `StrutAgentSession` plus one `StrutJob` node with an `IN_JOB` edge, upserted in the run-end projection hook. No `StrutArtifact` node.
To rule on. The plan lists seven choices under "Proposed rulings", none made yet. Continuing another job's thread from inside a turn is named and deliberately not designed.