Switch the extract and summarise stages on the page, over config.yaml

A stored `stage:extract` / `stage:summarise` in the state table overrides
extract.enabled / summarise.enabled; with none, the file applies. One
function, stages.stage_on, decides, and both tasks' finders ask it.
GET/PUT /api/stages and `app stages` show and set it with who would answer;
Runs has a "What the model reads" block, and a search agent's card says when
a stage is off and links to the switch.

Co-Authored-By: Claude Sonnet <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Sonnet committed 2026-09-29 15:54:53 +02:00
1 parent a629eda8de
commit 471ace703c
15 files changed
+593 -23

No files matched your search

+7
View File
@@ -111,6 +111,13 @@
</div>
<p class="lede">On the schedule below the pull reads every board, embed and score rank every role, extract fills in what the best postings say, score runs again with what it filled, and summarise writes the summaries.</p>
{# The two stages that ask a model, switched here rather than in config.yaml
(`GET /api/stages`); jobs.js draws the rows into #stages-list. #}
<h3 id="stages">What the model reads</h3>
<p class="lede">Extract and summarise ask a model, which costs tokens on the API. Turning one on starts nothing: the next pull runs it, and Run the pull now does so at once. How deep each reads is set per search agent, under <a href="/settings/agents">Search agents</a>.</p>
<div id="stages-list" class="srows"></div>
<p id="stages-answer" class="count" role="status"></p>
<h3>What is running</h3>
{# The element is the queue's own, served by its own router — see
pipeline/web/static/jobs.js; base.html loads its script. `idle="show"`