pagegraph 0.8.2 → 0.8.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -229,6 +229,8 @@ export default defineSeoConfig({
229
229
  opencode: {
230
230
  configDirectory: ".pagegraph/opencode",
231
231
  defaultModel: "openrouter/example/model",
232
+ // Optional per-completion budget in milliseconds; default 180000.
233
+ timeoutMs: 360_000,
232
234
  models: {
233
235
  "research.keywords": "openrouter/example/research-model",
234
236
  },
@@ -254,9 +256,23 @@ pagegraph research keywords \
254
256
  ```
255
257
 
256
258
  Every workflow starts from the selected graph and context files, asks the SEO agent to search
257
- Executor's live catalog and compose suitable tools, then evaluates its structured observations with
258
- Jev. Tool paths are discovered at runtime; the generated skills contain editable starter recipes,
259
- not a fixed integration list.
259
+ Executor's live catalog and inspect the discovered tools' argument schemas, then evaluates its
260
+ structured observations with Jev. Tool paths are discovered at runtime; the generated skills contain
261
+ editable starter recipes, not a fixed integration list.
262
+
263
+ `workflows.opencode.timeoutMs` sets the positive integer deadline in milliseconds for each OpenCode
264
+ completion (maximum `2147483647`). It covers prompt admission, model execution, and any workflow JSON
265
+ repair prompt. A later action or continuation turn gets a separate budget. The default is `180000`.
266
+
267
+ Selecting pages preserves their incoming and outgoing relationships to non-selected pages in
268
+ `evidence.neighborhood`. Those neighbors supply architecture context without becoming workflow
269
+ targets or consuming `--limit`. The agent receives the limit before research, and PageGraph also caps
270
+ the decoded result before decisions.
271
+
272
+ After research is validated, the run directory contains `research.json`: an incomplete checkpoint
273
+ with the collected evidence, decision inputs, and OpenCode transcript. A downstream failure reports
274
+ its path so the evidence remains available for diagnosis. Only a completed workflow writes
275
+ `run.json` and `summary.md`; a research checkpoint is not a final recommendation.
260
276
 
261
277
  | Command | Result |
262
278
  | --- | --- |