@c4a/context-cli 0.7.0 → 0.7.4

package/providers/context/skills/run-indexer-agent-step/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: context-run-indexer-agent-step
-description: Execute one Context-authorized Indexer workset using only the current Route input and resolved Provider instructions.
+description: Execute one Context-authorized Indexer workset using only the current Route input, resolved Provider instructions, and authorized workset View.
 metadata:
   agent-graph: path:../../provider.yaml
   agent-graph.graph: indexer
@@ -10,11 +10,37 @@ metadata:
 # Run one Indexer Agent step
 
 Read the current Route action input and every required Resource marked `read-required`.
-Use only the materialized `resolved-indexer-instructions` content, the supplied workset,
-composition input, and Context-authorized evidence. Do not discover another Skill, read a
-Provider path, widen source scope, run a program, or invent missing evidence.
+Use only the materialized `resolved-indexer-instructions`, the supplied workset authority,
+and the single `authorized-indexer-workset-view`. The View already contains the exact facts,
+dependencies, Provider fragments, and Author authority authorized for this workset. Do not
+construct evidence read requests, manage pages or cursors, discover another Skill, read a
+Provider path, widen source scope, run an unauthorized Indexer or parser, or invent missing
+evidence.
 
-Return exactly `context.indexer.agent-step-result/v1`. Preserve the supplied input and
-execution request digests, bind the materialized instruction payload digest, and include
-the current Host execution receipt. Context performs the final schema, evidence, owner,
-scope, and workset validation; never report success merely because the prose looks valid.
+Author the stage-specific operation Result described by the Action output schema. Copy dynamic
+authority values only from the Route input or authorized View; in particular,
+`consumed_input_view_digest` means `run_request.composition_input.view_digest`, not the
+authorized View's own digest.
+
+For a Partition Result, read `partition-authority` before choosing each group subject. A string
+changes only the local key and preserves `base_subject_key` namespace and kind. Use an explicit
+subject object only when its kind is listed in `subject_key_contract.kinds`; never invent a kind
+from a page title, operation name, framework term, or business vocabulary.
+
+For an Author Result, a selected parser Fact is an immutable projection, not material for a new
+Fact shape. Build it only from the matching View items:
+
+- `fact_ref` = the `fact` item's `value.fact_ref`;
+- `fact_kind` = the `fact` item's `value.kind`;
+- `subject_key` = `author-authority.expected_subject_key`;
+- `value` = the `fact` item's `value.payload` exactly, without wrapping, renaming, or adding fields;
+- `evidence_refs` = the source-span evidence refs reached through the matching `selected-fact`
+  dependency node.
+
+Build each EvidenceBinding from that same source-span dependency's `evidence_ref`, `source_ref`,
+`module_ref`, `locator`, and `content_digest`. Do not infer a new locator or digest from prose.
+Do not invent or manually copy read-receipt, execution-receipt, or stable-result digests. The Host
+integration must use Context's protocol builders to derive content-addressed fields and finalize
+exactly `context.indexer.agent-step-result/v1`; Context performs the existing dependency, schema,
+owner, scope, and workset validation before reporting success. Never report success merely because
+the envelope or prose looks valid.

package/providers/context/skills/run-indexer-lifecycle/SKILL.md

@@ -0,0 +1,37 @@
+---
+name: context-run-indexer-lifecycle
+description: Continue the sole Context registry-and-Provider indexing route; use only when the workspace Graph selects route.indexer.lifecycle-required.
+metadata:
+  agent-graph: path:../../provider.yaml
+  agent-graph.graph: workspace
+  agent-graph.entry: context
+---
+
+# Run The Indexer Lifecycle
+
+This Action continues the sole registry-and-Provider indexing route. Use the
+current Indexer subroute for semantic partitioning, structure review, authoring,
+composition, layout, Candidate compilation, and repair. Existing approved
+knowledge is an input to incremental planning, never a parallel authoring path.
+
+Read
+`node_modules/@c4a/context/docs/guides/indexer-provider-and-customization.md`,
+then follow the root `context` Skill's Indexer instructions and the exact
+structured outcomes returned by `context indexer ...`. Start by forming and
+confirming requirements when `src/indexers.yaml` is absent or stale; otherwise
+continue the current registry selection, Provider resolution, execution,
+reconciliation, mechanical validation, layout, and Candidate compile subroute. The Indexer
+Graph is the authority for each substep. Never skip directly to compile and
+never synthesize a Provider Result or a default plan.
+
+Stop on the first Host permission, non-delegatable Gate, validation failure, or
+human decision. Once current Candidates have been compiled, return control to
+`context status --format json`; the workspace Graph will select Knowledge
+Review. The Action output is only:
+
+```json
+{
+  "protocol": "context.indexer.lifecycle-continuation/v1",
+  "outcome": "follow-indexer-route"
+}
+```

package/providers/context/skills/run-indexer-post-author-composer/SKILL.md

@@ -9,13 +9,42 @@ metadata:
 
 # Run one post-author composer
 
-Read the current Route input and the exact `context.indexer.layer-fragment-request/v1`.
-Use only its bounded `primary_result_view`, selected `composer_ref`, allowed target refs,
-and the resolved instructions supplied by Context. Do not discover another Skill or
-composer, read project files outside the request, change a SubjectKey, create a complete
-primary Result, or widen the target set.
+Read both required Route resources before acting:
 
-Return exactly `context.indexer.layer-fragment-result/v1`. Preserve the request digest,
-composer ref, and consumed PrimaryResultView digest. Return only zero or more
-`derived-artifact-proposal` fragments for allowed targets. Context performs final schema,
-identity, evidence, policy, layer, and receipt validation before accepting the Result.
+- `resolved-indexer-instructions` defines the selected Composer's job;
+- `authorized-indexer-workset-view` is the complete, bounded PrimaryResult View.
+
+Use only the target aliases and source aliases exposed by that View and the current Route.
+Do not discover another Skill or Composer, scan project files outside the View, change a
+SubjectKey, rewrite the primary Result, or widen the target set.
+
+Return only the minimal semantic JSON accepted by the current Route:
+
+```json
+{
+  "stage": "post-author",
+  "outcome": "complete",
+  "proposals": [
+    {
+      "target": "target:1",
+      "artifact_kind": "content",
+      "title": "Reader-facing title",
+      "summary": "Why this derived page is useful",
+      "sections": [
+        {
+          "key": "overview",
+          "heading": "Overview",
+          "markdown": "Reader-facing content",
+          "source_refs": ["fact:1"]
+        }
+      ]
+    }
+  ],
+  "diagnostics": []
+}
+```
+
+If no derived Artifact is needed, return `outcome: "complete"` with an empty
+`proposals` array. If the bounded inputs cannot be processed, return `outcome: "failed"`
+with at least one diagnostic. Context expands aliases, validates evidence and policy,
+creates internal fragments and digests, then resumes the same production state machine.

package/docs/document-optimization.md

@@ -1,82 +0,0 @@
-# Document revisions
-
-New workspaces enable source-constrained editorial revision by default. It is a
-build phase for source-backed prose that improves publication value and
-readability without changing the approved knowledge page or inventing facts.
-
-To opt out during initialization:
-
-```bash
-context init context --no-optimize-docs
-```
-
-Existing workspaces retain their current setting. Change it explicitly with:
-
-```bash
-context optimize-docs enable
-context optimize-docs disable
-context status --format json
-```
-
-When enabled, the workflow plans only new or changed source Sections. For each
-Section the CLI reports mechanical signals and allowed actions. The Agent may
-keep it, repair local presentation, reshape it within the same source boundary,
-or omit it when the CLI confirms a non-knowledge reason. Only changed pages
-receive a full revision beside their approved page:
-
-Signals are review leads rather than a complete readability verdict. The Agent
-must read every Section. Keeping a signaled Section requires a concrete,
-decision-only assessment explaining why each signal is a false positive or why
-editing would reduce source fidelity. The assessment names every reported
-signal code so the CLI can verify complete coverage; it is never published.
-Time, cost, workload, batch size, or delivery pressure are not valid keep or
-defer reasons. Fully managed operation applies all safe repairs, reshapes, and
-eligible omissions autonomously; only genuinely unavailable semantic input is
-batched for the user. A large fragment set must meet the same quality standard
-as a single fragment.
-
-```text
-knowledge/guides/setup.md
-knowledge/guides/setup__revision.md
-```
-
-The `__revision.md` suffix is reserved. Default knowledge discovery, structure,
-search, and package selection exclude revision files. Validation and build
-associate a revision with its sibling base page and apply it under the original
-package path. A revision stores only the base digest that cannot be inferred;
-its path and revised Sections are derived. Unchanged Sections inside a full
-revision are inferred. A page with no changes stores only one derived negative
-cache key below `.tmp/context-runtime/document-optimization/`; replacement
-prose and Section metadata are not duplicated there.
-
-Typical repair candidates are Markdown syntax, spacing, empty table rows, and
-descriptive labels for links whose purpose is already present. Reshape handles
-wide tables, long cells, mixed images and links, or long prose by creating a
-clearer structure inside the same Section. Unanswered question sets, empty
-placeholders, decision-free drafts, duplicates, and obsolete-only Sections may
-be omitted. Questions with answers, limitations with an action, and
-deprecations with a replacement remain knowledge. Link targets, images, code,
-commands, numbers, identifiers, conditions, and source references are protected
-through validation. Ambiguous currency, ownership, or sensitive values appear
-together in the plan's `input_requests` and require one batched user decision.
-
-After capture, review, or build, a user may ask conversationally to correct one
-existing knowledge page. Start the correction by title, approved path, or
-ViewRef:
-
-```bash
-context revise "<title, approved path, or ViewRef>" --format json
-```
-
-Context starts `route.document-revision.requested` only after resolving one
-page. Ambiguous requests return candidates instead of guessing. Edit
-reader-visible prose within existing `context:section` boundaries without
-changing provenance or protected values, then run `context optimize-docs
-validate`. A valid correction ends
-the request and the next Route offers a package build when output is stale. If
-broad document optimization was disabled, this entry activates only the target
-page and keeps every other eligible page out of the pending batch. If upstream
-content changes, the revision becomes a conflict and must be reviewed again.
-
-Disabling moves active revision pages to Context runtime recovery storage. The
-next build uses approved knowledge directly.

package/docs/document-optimization.zh-CN.md

@@ -1,70 +0,0 @@
-# 文档修订页
-
-新工作区默认开启来源约束的编辑修订。它是面向来源可追溯正文的构建阶段，在不修改
-正式知识页、不补写事实的前提下，提高发布内容的知识价值和可读性。
-
-初始化时如需关闭：
-
-```bash
-context init context --no-optimize-docs
-```
-
-已有工作区保持当前设置，可显式调整：
-
-```bash
-context optimize-docs enable
-context optimize-docs disable
-context status --format json
-```
-
-开启后，工作流只规划新增或发生变化的来源 Section。CLI 为每个 Section 枚举机械
-信号和允许动作；Agent 可以保持不变、局部修复、在同一来源边界内重排，或在 CLI
-确认其属于无知识价值内容时省略。只有实际变化的页面才在原文旁生成完整修订页：
-
-```text
-knowledge/guides/setup.md
-knowledge/guides/setup__revision.md
-```
-
-`__revision.md` 是保留后缀。默认知识发现、结构、检索和包选择都排除修订页；
-校验和构建根据文件名把它关联回同目录原文，并仍按原文件名输出。修订页只额外
-记录无法推导的原文摘要；原文路径和修订 Section 由文件名与正文差异推导。完整修订
-页中未变化的 Section 直接推导为 `keep`；整页都不修改时，运行时只保存一个页级负
-缓存键，不重复保存改写正文或 Section 元数据。
-
-局部修复适用于 Markdown、空白、空表格行，以及来源已经说明用途的链接名称。结构
-重排适用于宽表格、长单元格、图片与链接混排或过长段落，但不能跨 Section 搬运事实。
-未回答的问题集合、空占位、没有结论的草稿、重复内容和只有旧稿的 Section 可以省略；
-有答案的问句、有影响和动作的限制、带替代方案的废弃说明仍属于知识。链接目标、图片、
-代码、命令、数字、标识符、条件和来源引用由 CLI 机械保护。事实时效、职责或敏感值不
-明确时，CLI 会把相关 Section 集中列入计划的 `input_requests`，工作流只询问用户一次，
-不会自动省略。
-
-机械信号只是审阅线索，不代表最终结论，也不是完整的可读性检查。Agent 必须阅读每个
-Section；若对带信号的 Section 选择 `keep`，需要在本次决策中逐段说明信号为何误报，
-或修改为何会损害来源忠实度，并在说明中逐一写出对应的信号 code，供 CLI 机械核对覆盖
-是否完整。CLI 会拒绝没有具体说明或漏掉信号的 `keep`，但这段说明只用于当前决策审计，
-不会写入正式知识、修订页或构建产物。部分信号误报也不能成为整批保持不变的理由。
-
-时间、成本、工作量、批次规模或赶进度都不是 `keep`、跳过或延期的有效理由。全托管
-模式下，Agent 应自动完成所有安全的局部修复、结构重排和允许省略，直到当前批次收敛；
-只有确实无法从当前 Section 或来源证据取得的语义信息，才集中向用户询问一次。片段多
-不能降低质量标准，也不能把整批默认判为 `keep`。
-
-工作区采集、审核或构建完成后，用户可以直接通过对话要求修正某篇知识。Agent
-先使用标题、正式知识路径或 ViewRef 启动修订：
-
-```bash
-context revise "<标题、正式知识路径或 ViewRef>" --format json
-```
-
-CLI 只在目标唯一时创建或复用该页修订，并由 Agent Graph 返回
-`route.document-revision.requested`。若有多个候选，Agent 必须依据当前对话选定，
-否则询问用户。Agent 只在已有 `context:section` 内编辑面向读者的正文，不修改来源
-信息、Section 边界和受保护值，随后运行 `context optimize-docs validate`。校验成功后，
-请求自动结束；若产物因此过期，下一条 Route 会直接提示重新构建。即使工作区
-此前关闭了整库文档编译优化，这个入口也只激活目标页面，不会让其他页面进入
-待优化队列。上游内容变化后，旧修订页会成为冲突，不会被静默应用。
-
-关闭后，修订页会被移动到 Context 的运行时恢复目录，下一次构建直接使用正式
-知识。

package/indexers/capability-manifest.json

@@ -1,40 +0,0 @@
-{
-  "protocol": "context.indexer.release-capability-manifest/v1",
-  "package": "@c4a/context-cli",
-  "version": "0.7.0",
-  "channel": "latest",
-  "dist_tag": "latest",
-  "capabilities": [
-    {
-      "id": "host-action-v2",
-      "state": "ready",
-      "required_milestone": "0.7.0-preview.1"
-    },
-    {
-      "id": "provider-lifecycle",
-      "state": "ready",
-      "required_milestone": "0.7.0-preview.1"
-    },
-    {
-      "id": "code-indexer",
-      "state": "ready",
-      "required_milestone": "0.7.0-preview.2"
-    },
-    {
-      "id": "markdown-indexer",
-      "state": "ready",
-      "required_milestone": "0.7.0-preview.3"
-    },
-    {
-      "id": "enterprise-overlay-trust",
-      "state": "ready",
-      "required_milestone": "0.7.0-rc.1"
-    },
-    {
-      "id": "phase-g-cutover",
-      "state": "ready",
-      "required_milestone": "0.7.0"
-    }
-  ],
-  "manifest_digest": "sha256:308607b2f4cd11fbab4e4d6596f00f633e2aed0e60af24f44f0c3fb63efd6c08"
-}