@c4a/context 0.6.1-beta.2 → 0.6.1-beta.3

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.
@@ -17,6 +17,12 @@ Use `--dev` when testing a local link or prepared pack before the matching SDK
17
17
  version is published. It writes a `file:` dependency to the SDK resolved beside
18
18
  the active CLI. Published users can omit it and install the versioned SDK from
19
19
  the registry.
20
+
21
+ Without `project-dir`, init uses the dedicated `context/` directory. Initializing
22
+ inside a non-empty directory that is not already a Context workspace is blocked
23
+ before any files are written; use the returned `--allow-nonempty` command only
24
+ after confirming that the existing files should share the workspace root.
25
+
20
26
  When operating through an Agent plugin, use the installed Context continuation
21
27
  entry from the project root after initialization. It consumes
22
28
  `workflow.current`, loads only the selected resources, and calls lower-level
@@ -30,6 +30,14 @@ without another question. Review uses the CLI's atomic `context review
30
30
  approve-all ... --managed` route; valid structure staging records
31
31
  `confirmed_by: managed-session`.
32
32
 
33
+ After the first managed status evaluation, use
34
+ `context run --managed --until blocked-or-complete --format json` when the
35
+ current work can advance through consecutive mechanical routes. The CLI
36
+ executes only a unique immediate non-read command, re-evaluates after each
37
+ receipt, and stops before Agent interpretation, configuration, missing
38
+ authority, diagnostics, or multiple commands. Continue from the returned
39
+ `workflow.current`.
40
+
33
41
  This is execution authority, not project configuration. Do not add it to
34
42
  `defineProject`, environment files, or committed workspace state, and do not
35
43
  carry it into a new conversation. It never grants a new source boundary or
@@ -372,6 +380,7 @@ context run align:<type>:<source>:<collection> --view read-plan --format json
372
380
  context run align:<type>:<source>:<collection> --view source-index --compact --format json
373
381
  context run align:<type>:<source>:<collection> --view span-detail --span <source-ref> --format json
374
382
  context run align:<type>:<source>:<collection> --view span-text --span <source-ref> --format json
383
+ context run align:<type>:<source>:<collection> --view existing-knowledge --query <title-or-stable-ref> --format json
375
384
  context run align:<type>:<source>:<collection> --view schema --format json
376
385
  context run align:<type>:<source>:<collection> --validate --input <structure.yaml> --format json
377
386
  context run align:<type>:<source>:<collection> --view structure-summary --input <structure.yaml> --format json
@@ -380,9 +389,12 @@ context run align:<type>:<source>:<collection> --stage --input <structure.yaml>
380
389
 
381
390
  Read source material only through these evidence views. `source-index` gives a
382
391
  compact refs-first map when run with `--compact`; use `span-detail` /
383
- `span-text` only for exact evidence. The CLI applies deterministic boundary
384
- repairs internally and returns only blockers that need Agent judgment. For
385
- oversized Views, use the returned structural diagnostics while classifying
392
+ `span-text` only for exact evidence. Before introducing a new Node identity,
393
+ use the targeted `existing-knowledge` View returned by the read plan to inspect
394
+ approved stable refs; do not inspect `knowledge/**` directly. The CLI applies
395
+ deterministic boundary repairs internally and returns only blockers that need
396
+ Agent judgment. For oversized Views, use the returned structural diagnostics
397
+ while classifying
386
398
  child Nodes from evidence. Stage only after validation state is `ready`; stage
387
399
  opens the final `structure-summary` report for the current Route's confirmation
388
400
  gate. Ask a separate structure-design question only when evidence supports
@@ -250,6 +250,17 @@ returned by `workflow.current`. Source boundaries and unread source bodies,
250
250
  external operations, payload validation, deterministic close, and verification
251
251
  errors are never bypassed.
252
252
 
253
+ For consecutive mechanical routes, the Agent may run:
254
+
255
+ ```bash
256
+ context run --managed --until blocked-or-complete --format json
257
+ ```
258
+
259
+ This is a bounded host loop over the same revisioned routes. It stops before
260
+ read-only interpretation, project configuration, unresolved authority,
261
+ diagnostics, or a non-unique command plan; it does not add another workflow
262
+ entry or make semantic decisions.
263
+
253
264
  ### `captureFile`
254
265
 
255
266
  Capture a registered file source into a committed normalized document snapshot.
@@ -354,6 +365,7 @@ context run align:file:<source-name>:architecture --view read-plan --format json
354
365
  context run align:file:<source-name>:architecture --view source-index --compact --format json
355
366
  context run align:file:<source-name>:architecture --view span-detail --span <source-ref> --format json
356
367
  context run align:file:<source-name>:architecture --view span-text --span <source-ref> --format json
368
+ context run align:file:<source-name>:architecture --view existing-knowledge --query <title-or-stable-ref> --format json
357
369
  context run align:file:<source-name>:architecture --view schema --format json
358
370
  context run align:file:<source-name>:architecture --view semantic-rules --format json
359
371
  context run align:file:<source-name>:architecture --validate --input <structure.yaml> --format json
@@ -371,6 +383,14 @@ schema and full reports stay behind explicit Views, while `--verbose` restores
371
383
  the full phase result and repeated contracts. Long diagnostics return a compact
372
384
  first page plus an exact diagnostics continuation command.
373
385
 
386
+ `existing-knowledge` is the authoring-time lookup for approved identities. It
387
+ returns stable NodeRefs, ViewRefs, titles, tags, collections, and section counts
388
+ without exposing workspace storage paths. `--query` performs deterministic
389
+ case-insensitive exact/prefix/substring matching; `--collection`,
390
+ `--node-type`, `--page-size`, and the returned continuation command narrow or
391
+ page the same View. Use it after reading source evidence and before introducing
392
+ a new Node identity. Structure validation remains the final duplicate gate.
393
+
374
394
  Align and compile evidence results include `semantic_rules`. Its `required`
375
395
  array is the rule subset selected for the current judgment, with a selection
376
396
  reason and content digest for each rule. `handle`, `digest`, and
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@c4a/context",
3
- "version": "0.6.1-beta.2",
3
+ "version": "0.6.1-beta.3",
4
4
  "type": "module",
5
5
  "description": "Context SDK — project-local configuration and workspace primitives",
6
6
  "license": "MIT",