@c4a/context-cli 0.6.7 → 0.6.9

package/plugins/claude/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
-  "name": "context",
-  "description": "Maintain a project-local knowledge workspace through init and next-step agent guidance.",
-  "version": "0.6.7",
+  "name": "c4a",
+  "description": "Start or continue a project-local knowledge workspace through one graph-routed entry.",
+  "version": "0.6.9",
   "author": {
     "name": "c4a"
   },

package/plugins/claude/commands/context.md

@@ -0,0 +1,121 @@
+---
+description: "Build or continue structured, traceable Agent knowledge from documents and code."
+argument-hint: "[project-dir or user intent]"
+allowed-tools: Bash(context:*), Bash(bun:*), Bash(cd *)
+---
+
+## Your Task
+
+Context is a knowledge management tool built for Agent knowledge workflows. It
+compiles Feishu/Lark documents, local Markdown, repository code, and manually
+curated business material into structured, traceable knowledge, then produces
+knowledge packages, LLM-ready documents, or Agent Skills. The CLI packages all
+workflow guidance, knowledge-building procedures, and code-indexing capabilities
+needed to produce that knowledge; follow its returned commands and resources
+for the next action.
+
+Use this as the single conversational entry for Context. Let the CLI locate an
+existing workspace, relocate into it, initialize a requested workspace, or
+evaluate its current workflow. Do not infer the workspace state yourself.
+
+### Enter the workspace
+
+Run:
+
+```bash
+context entry [project-dir] --language <language> --format json
+```
+
+Use the user's explicit language choice when present; otherwise pass `zh-CN`
+for a Chinese conversation and `en` for an English conversation. Pass
+`project-dir`, `--name`, `--dev`, or `--debug` only when the user explicitly
+requested that initialization choice. Pass `--managed` only after the user
+explicitly authorizes fully managed operation in this conversation.
+
+If the `context` process itself cannot start because the command is missing
+(`ENOENT`, or shell exit 127 explicitly identifying `context` as the missing
+command, such as `command not found: context` or `context: command not found`),
+explain that the global CLI is not installed and ask the user to install or
+authorize installation with:
+
+```bash
+npm install -g @c4a/context-cli@latest
+context plugin install
+```
+
+Stop after giving that recovery. Do not run an installation preflight, install
+automatically, or mistake a normal Context `not found` diagnostic for a missing
+executable.
+
+Execute only `next_action.command` returned by `context entry`:
+
+- `initialize-workspace` writes a new workspace. Execute it immediately only
+  when the user explicitly requested initialization through this entry;
+  otherwise explain the target root and ask for confirmation. Preserve the
+  `init-target-nonempty` confirmation.
+- `enter-workspace` and `evaluate-workflow` are read-only and need no additional
+  confirmation.
+- After initialization, execute the exact setup command returned by `context
+  init`, enter the project root, read the generated `AGENTS.md`, and run this
+  entry again.
+
+### Conversation modes
+
+Enable debugging only when the user explicitly requests it. Run `context debug
+enable` first; debugging records traces below `.tmp/context-runtime/debug/` but
+does not grant workflow authority or provide source evidence.
+
+For explicitly authorized fully managed operation, use:
+
+```bash
+context run --managed --until blocked-or-complete --format json
+```
+
+Use `--managed` for every resumed workflow evaluation in the same active
+request. Never persist or reuse that authority in another conversation, and
+stop using it when the conversation ends or the user revokes it. Pass any
+additional `--authority` only when the user explicitly grants that authority in
+this conversation.
+
+The managed loop executes only Route-selected work and returns the current
+`workflow.current` whenever Agent reading, project configuration, a human Gate,
+host execution, diagnostics, or a non-unique plan needs attention. Resume from
+that returned Route; never reconstruct a command from an earlier step.
+
+### Follow the current Route
+
+Treat `workflow.current` as the current-step authority:
+
+1. Read every `resources.required` item whose `read_state` is `read-required`.
+   Read a returned `path` completely, or execute a returned resource `command`
+   and read its complete output file. Materializing a resource is not reading
+   it. Keep the merged receipts only in this conversation and submit them with
+   the exact returned `context status --resource-receipts @<file>` command. Use
+   `after_read_receipts` only after the full resource has been read. The exact
+   `resources.after_read.command` already returns the re-evaluated
+   `workflow.current`; continue from it without an additional status call.
+2. At a Gate, keep inspection and resolution phase-local. Read an
+   `inspection_action` resource only while inspecting the decision, and read a
+   `resolution_action` resource only after the user confirms the Gate. Neither
+   replaces ordinary required resources.
+3. Execute only `commands` returned by the Route, preserving revision and
+   authority flags exactly. A command marked `after-human-confirmation` waits
+   for the current Gate decision. Run a command whose `execution.target` is `agent-host`
+   as an exact top-level host action with the required host access,
+   not inside a restricted child sandbox. Follow the Route-selected procedure
+   for its audit and approval contract; never invent a payload, destination, or
+   substitute command.
+4. If `configuration` is present, edit only the named project file and use the
+   selected resources as its contract.
+5. After every action or configuration change, run status again. The managed
+   loop performs this re-evaluation internally. A phase-local `next_action` can
+   continue that operation but never replaces the workspace Route.
+
+Explain, ask, confirm, and summarize in the user's current conversation
+language. Keep commands, flags, paths, ids, status values, JSONL keys,
+`source_ref` values, and copied CLI tokens unchanged.
+
+Do not infer repo sources, extraction scope, review decisions, or package output
+choices from surrounding files. Do not call source-repo operations such as
+clone, checkout, reset, fetch, install, build, or test without explicit user
+approval.

package/plugins/codex/.codex-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
-  "name": "context",
-  "version": "0.6.7",
-  "description": "Maintain a project-local knowledge workspace through init and next-step agent guidance.",
+  "name": "c4a",
+  "version": "0.6.9",
+  "description": "Start or continue a project-local knowledge workspace through one graph-routed entry.",
   "author": {
     "name": "c4a"
   },
@@ -18,7 +18,7 @@
   "skills": "./skills/",
   "interface": {
     "displayName": "C4A Context",
-    "shortDescription": "Initialize and advance a local, source-linked project knowledge workspace.\nv0.6.7",
+    "shortDescription": "Initialize and advance a local, source-linked project knowledge workspace.\nv0.6.9",
     "longDescription": "Create a Context workspace and use agent-guided next steps to register sources, run extraction, review candidates, build package outputs, and verify health without silently mutating source repositories.",
     "developerName": "c4a",
     "category": "Productivity",

package/plugins/codex/skills/context/SKILL.md

@@ -0,0 +1,133 @@
+---
+name: context
+description: >
+  Build or continue structured, traceable Agent knowledge from documents and code. Use the local `context` CLI for workspace writes.
+tools:
+  - Bash
+---
+
+# Context
+
+Public Context entry for agents that expose skills instead of slash commands.
+
+- Public entry: `context`
+- Host command names are installation-specific; use the command or skill surfaced by the current host instead of deriving a filesystem path from an example name.
+- CLI primitive prefix: `context ...`
+
+---
+
+## Workflow
+
+Context is a knowledge management tool built for Agent knowledge workflows. It
+compiles Feishu/Lark documents, local Markdown, repository code, and manually
+curated business material into structured, traceable knowledge, then produces
+knowledge packages, LLM-ready documents, or Agent Skills. The CLI packages all
+workflow guidance, knowledge-building procedures, and code-indexing capabilities
+needed to produce that knowledge; follow its returned commands and resources
+for the next action.
+
+Use this as the single conversational entry for Context. Let the CLI locate an
+existing workspace, relocate into it, initialize a requested workspace, or
+evaluate its current workflow. Do not infer the workspace state yourself.
+
+### Enter the workspace
+
+Run:
+
+```bash
+context entry [project-dir] --language <language> --format json
+```
+
+Use the user's explicit language choice when present; otherwise pass `zh-CN`
+for a Chinese conversation and `en` for an English conversation. Pass
+`project-dir`, `--name`, `--dev`, or `--debug` only when the user explicitly
+requested that initialization choice. Pass `--managed` only after the user
+explicitly authorizes fully managed operation in this conversation.
+
+If the `context` process itself cannot start because the command is missing
+(`ENOENT`, or shell exit 127 explicitly identifying `context` as the missing
+command, such as `command not found: context` or `context: command not found`),
+explain that the global CLI is not installed and ask the user to install or
+authorize installation with:
+
+```bash
+npm install -g @c4a/context-cli@latest
+context plugin install
+```
+
+Stop after giving that recovery. Do not run an installation preflight, install
+automatically, or mistake a normal Context `not found` diagnostic for a missing
+executable.
+
+Execute only `next_action.command` returned by `context entry`:
+
+- `initialize-workspace` writes a new workspace. Execute it immediately only
+  when the user explicitly requested initialization through this entry;
+  otherwise explain the target root and ask for confirmation. Preserve the
+  `init-target-nonempty` confirmation.
+- `enter-workspace` and `evaluate-workflow` are read-only and need no additional
+  confirmation.
+- After initialization, execute the exact setup command returned by `context
+  init`, enter the project root, read the generated `AGENTS.md`, and run this
+  entry again.
+
+### Conversation modes
+
+Enable debugging only when the user explicitly requests it. Run `context debug
+enable` first; debugging records traces below `.tmp/context-runtime/debug/` but
+does not grant workflow authority or provide source evidence.
+
+For explicitly authorized fully managed operation, use:
+
+```bash
+context run --managed --until blocked-or-complete --format json
+```
+
+Use `--managed` for every resumed workflow evaluation in the same active
+request. Never persist or reuse that authority in another conversation, and
+stop using it when the conversation ends or the user revokes it. Pass any
+additional `--authority` only when the user explicitly grants that authority in
+this conversation.
+
+The managed loop executes only Route-selected work and returns the current
+`workflow.current` whenever Agent reading, project configuration, a human Gate,
+host execution, diagnostics, or a non-unique plan needs attention. Resume from
+that returned Route; never reconstruct a command from an earlier step.
+
+### Follow the current Route
+
+Treat `workflow.current` as the current-step authority:
+
+1. Read every `resources.required` item whose `read_state` is `read-required`.
+   Read a returned `path` completely, or execute a returned resource `command`
+   and read its complete output file. Materializing a resource is not reading
+   it. Keep the merged receipts only in this conversation and submit them with
+   the exact returned `context status --resource-receipts @<file>` command. Use
+   `after_read_receipts` only after the full resource has been read. The exact
+   `resources.after_read.command` already returns the re-evaluated
+   `workflow.current`; continue from it without an additional status call.
+2. At a Gate, keep inspection and resolution phase-local. Read an
+   `inspection_action` resource only while inspecting the decision, and read a
+   `resolution_action` resource only after the user confirms the Gate. Neither
+   replaces ordinary required resources.
+3. Execute only `commands` returned by the Route, preserving revision and
+   authority flags exactly. A command marked `after-human-confirmation` waits
+   for the current Gate decision. Run a command whose `execution.target` is `agent-host`
+   as an exact top-level host action with the required host access,
+   not inside a restricted child sandbox. Follow the Route-selected procedure
+   for its audit and approval contract; never invent a payload, destination, or
+   substitute command.
+4. If `configuration` is present, edit only the named project file and use the
+   selected resources as its contract.
+5. After every action or configuration change, run status again. The managed
+   loop performs this re-evaluation internally. A phase-local `next_action` can
+   continue that operation but never replaces the workspace Route.
+
+Explain, ask, confirm, and summarize in the user's current conversation
+language. Keep commands, flags, paths, ids, status values, JSONL keys,
+`source_ref` values, and copied CLI tokens unchanged.
+
+Do not infer repo sources, extraction scope, review decisions, or package output
+choices from surrounding files. Do not call source-repo operations such as
+clone, checkout, reset, fetch, install, build, or test without explicit user
+approval.

package/plugins/cursor/.cursor-plugin/plugin.json

@@ -1,8 +1,8 @@
 {
-  "name": "context",
+  "name": "c4a",
   "displayName": "C4A Context",
-  "version": "0.6.7",
-  "description": "Maintain a project-local knowledge workspace through init and next-step agent guidance.",
+  "version": "0.6.9",
+  "description": "Start or continue a project-local knowledge workspace through one graph-routed entry.",
   "author": {
     "name": "Context4AI",
     "email": "support@context4ai.dev"

package/plugins/cursor/README.md

@@ -7,7 +7,7 @@ Edit `packages/context-cli/plugin/` and rerun the build instead.
 Install shape:
 
 - Marketplace/GitHub plugin shape: this directory is a plugin root with `.cursor-plugin/plugin.json` and `commands/`.
-- User entries are under `commands/`; Cursor command files are prefixed as `context-*` to avoid global slash-command collisions.
-- Local plugin fallback: symlink or copy this directory to `~/.cursor/plugins/local/context/` only when Marketplace import is unavailable.
+- User entries are under `commands/`; Cursor command files are prefixed as `c4a-*` to avoid global slash-command collisions.
+- Local plugin fallback: symlink or copy this directory to `~/.cursor/plugins/local/c4a/` only when Marketplace import is unavailable.
 
 Cursor Marketplace installs this plugin root directly.

package/plugins/cursor/commands/c4a-context.md

@@ -0,0 +1,125 @@
+---
+description: "Build or continue structured, traceable Agent knowledge from documents and code."
+argument-hint: "[project-dir or user intent]"
+allowed-tools: Bash(context:*), Bash(bun:*), Bash(cd *)
+---
+
+Start or continue a C4A Context knowledge workspace.
+
+---
+
+## Workflow
+
+Context is a knowledge management tool built for Agent knowledge workflows. It
+compiles Feishu/Lark documents, local Markdown, repository code, and manually
+curated business material into structured, traceable knowledge, then produces
+knowledge packages, LLM-ready documents, or Agent Skills. The CLI packages all
+workflow guidance, knowledge-building procedures, and code-indexing capabilities
+needed to produce that knowledge; follow its returned commands and resources
+for the next action.
+
+Use this as the single conversational entry for Context. Let the CLI locate an
+existing workspace, relocate into it, initialize a requested workspace, or
+evaluate its current workflow. Do not infer the workspace state yourself.
+
+### Enter the workspace
+
+Run:
+
+```bash
+context entry [project-dir] --language <language> --format json
+```
+
+Use the user's explicit language choice when present; otherwise pass `zh-CN`
+for a Chinese conversation and `en` for an English conversation. Pass
+`project-dir`, `--name`, `--dev`, or `--debug` only when the user explicitly
+requested that initialization choice. Pass `--managed` only after the user
+explicitly authorizes fully managed operation in this conversation.
+
+If the `context` process itself cannot start because the command is missing
+(`ENOENT`, or shell exit 127 explicitly identifying `context` as the missing
+command, such as `command not found: context` or `context: command not found`),
+explain that the global CLI is not installed and ask the user to install or
+authorize installation with:
+
+```bash
+npm install -g @c4a/context-cli@latest
+context plugin install
+```
+
+Stop after giving that recovery. Do not run an installation preflight, install
+automatically, or mistake a normal Context `not found` diagnostic for a missing
+executable.
+
+Execute only `next_action.command` returned by `context entry`:
+
+- `initialize-workspace` writes a new workspace. Execute it immediately only
+  when the user explicitly requested initialization through this entry;
+  otherwise explain the target root and ask for confirmation. Preserve the
+  `init-target-nonempty` confirmation.
+- `enter-workspace` and `evaluate-workflow` are read-only and need no additional
+  confirmation.
+- After initialization, execute the exact setup command returned by `context
+  init`, enter the project root, read the generated `AGENTS.md`, and run this
+  entry again.
+
+### Conversation modes
+
+Enable debugging only when the user explicitly requests it. Run `context debug
+enable` first; debugging records traces below `.tmp/context-runtime/debug/` but
+does not grant workflow authority or provide source evidence.
+
+For explicitly authorized fully managed operation, use:
+
+```bash
+context run --managed --until blocked-or-complete --format json
+```
+
+Use `--managed` for every resumed workflow evaluation in the same active
+request. Never persist or reuse that authority in another conversation, and
+stop using it when the conversation ends or the user revokes it. Pass any
+additional `--authority` only when the user explicitly grants that authority in
+this conversation.
+
+The managed loop executes only Route-selected work and returns the current
+`workflow.current` whenever Agent reading, project configuration, a human Gate,
+host execution, diagnostics, or a non-unique plan needs attention. Resume from
+that returned Route; never reconstruct a command from an earlier step.
+
+### Follow the current Route
+
+Treat `workflow.current` as the current-step authority:
+
+1. Read every `resources.required` item whose `read_state` is `read-required`.
+   Read a returned `path` completely, or execute a returned resource `command`
+   and read its complete output file. Materializing a resource is not reading
+   it. Keep the merged receipts only in this conversation and submit them with
+   the exact returned `context status --resource-receipts @<file>` command. Use
+   `after_read_receipts` only after the full resource has been read. The exact
+   `resources.after_read.command` already returns the re-evaluated
+   `workflow.current`; continue from it without an additional status call.
+2. At a Gate, keep inspection and resolution phase-local. Read an
+   `inspection_action` resource only while inspecting the decision, and read a
+   `resolution_action` resource only after the user confirms the Gate. Neither
+   replaces ordinary required resources.
+3. Execute only `commands` returned by the Route, preserving revision and
+   authority flags exactly. A command marked `after-human-confirmation` waits
+   for the current Gate decision. Run a command whose `execution.target` is `agent-host`
+   as an exact top-level host action with the required host access,
+   not inside a restricted child sandbox. Follow the Route-selected procedure
+   for its audit and approval contract; never invent a payload, destination, or
+   substitute command.
+4. If `configuration` is present, edit only the named project file and use the
+   selected resources as its contract.
+5. After every action or configuration change, run status again. The managed
+   loop performs this re-evaluation internally. A phase-local `next_action` can
+   continue that operation but never replaces the workspace Route.
+
+Explain, ask, confirm, and summarize in the user's current conversation
+language. Keep commands, flags, paths, ids, status values, JSONL keys,
+`source_ref` values, and copied CLI tokens unchanged.
+
+Do not infer repo sources, extraction scope, review decisions, or package output
+choices from surrounding files. Do not call source-repo operations such as
+clone, checkout, reset, fetch, install, build, or test without explicit user
+approval.

package/plugins/skills/c4a-context/SKILL.md

@@ -0,0 +1,133 @@
+---
+name: c4a-context
+description: >
+  Build or continue structured, traceable Agent knowledge from documents and code. Use the local `context` CLI for workspace writes.
+tools:
+  - Bash
+---
+
+# Context
+
+Public Context entry for agents that expose skills instead of slash commands.
+
+- Public entry: `c4a-context`
+- Host command names are installation-specific; use the command or skill surfaced by the current host instead of deriving a filesystem path from an example name.
+- CLI primitive prefix: `context ...`
+
+---
+
+## Workflow
+
+Context is a knowledge management tool built for Agent knowledge workflows. It
+compiles Feishu/Lark documents, local Markdown, repository code, and manually
+curated business material into structured, traceable knowledge, then produces
+knowledge packages, LLM-ready documents, or Agent Skills. The CLI packages all
+workflow guidance, knowledge-building procedures, and code-indexing capabilities
+needed to produce that knowledge; follow its returned commands and resources
+for the next action.
+
+Use this as the single conversational entry for Context. Let the CLI locate an
+existing workspace, relocate into it, initialize a requested workspace, or
+evaluate its current workflow. Do not infer the workspace state yourself.
+
+### Enter the workspace
+
+Run:
+
+```bash
+context entry [project-dir] --language <language> --format json
+```
+
+Use the user's explicit language choice when present; otherwise pass `zh-CN`
+for a Chinese conversation and `en` for an English conversation. Pass
+`project-dir`, `--name`, `--dev`, or `--debug` only when the user explicitly
+requested that initialization choice. Pass `--managed` only after the user
+explicitly authorizes fully managed operation in this conversation.
+
+If the `context` process itself cannot start because the command is missing
+(`ENOENT`, or shell exit 127 explicitly identifying `context` as the missing
+command, such as `command not found: context` or `context: command not found`),
+explain that the global CLI is not installed and ask the user to install or
+authorize installation with:
+
+```bash
+npm install -g @c4a/context-cli@latest
+context plugin install
+```
+
+Stop after giving that recovery. Do not run an installation preflight, install
+automatically, or mistake a normal Context `not found` diagnostic for a missing
+executable.
+
+Execute only `next_action.command` returned by `context entry`:
+
+- `initialize-workspace` writes a new workspace. Execute it immediately only
+  when the user explicitly requested initialization through this entry;
+  otherwise explain the target root and ask for confirmation. Preserve the
+  `init-target-nonempty` confirmation.
+- `enter-workspace` and `evaluate-workflow` are read-only and need no additional
+  confirmation.
+- After initialization, execute the exact setup command returned by `context
+  init`, enter the project root, read the generated `AGENTS.md`, and run this
+  entry again.
+
+### Conversation modes
+
+Enable debugging only when the user explicitly requests it. Run `context debug
+enable` first; debugging records traces below `.tmp/context-runtime/debug/` but
+does not grant workflow authority or provide source evidence.
+
+For explicitly authorized fully managed operation, use:
+
+```bash
+context run --managed --until blocked-or-complete --format json
+```
+
+Use `--managed` for every resumed workflow evaluation in the same active
+request. Never persist or reuse that authority in another conversation, and
+stop using it when the conversation ends or the user revokes it. Pass any
+additional `--authority` only when the user explicitly grants that authority in
+this conversation.
+
+The managed loop executes only Route-selected work and returns the current
+`workflow.current` whenever Agent reading, project configuration, a human Gate,
+host execution, diagnostics, or a non-unique plan needs attention. Resume from
+that returned Route; never reconstruct a command from an earlier step.
+
+### Follow the current Route
+
+Treat `workflow.current` as the current-step authority:
+
+1. Read every `resources.required` item whose `read_state` is `read-required`.
+   Read a returned `path` completely, or execute a returned resource `command`
+   and read its complete output file. Materializing a resource is not reading
+   it. Keep the merged receipts only in this conversation and submit them with
+   the exact returned `context status --resource-receipts @<file>` command. Use
+   `after_read_receipts` only after the full resource has been read. The exact
+   `resources.after_read.command` already returns the re-evaluated
+   `workflow.current`; continue from it without an additional status call.
+2. At a Gate, keep inspection and resolution phase-local. Read an
+   `inspection_action` resource only while inspecting the decision, and read a
+   `resolution_action` resource only after the user confirms the Gate. Neither
+   replaces ordinary required resources.
+3. Execute only `commands` returned by the Route, preserving revision and
+   authority flags exactly. A command marked `after-human-confirmation` waits
+   for the current Gate decision. Run a command whose `execution.target` is `agent-host`
+   as an exact top-level host action with the required host access,
+   not inside a restricted child sandbox. Follow the Route-selected procedure
+   for its audit and approval contract; never invent a payload, destination, or
+   substitute command.
+4. If `configuration` is present, edit only the named project file and use the
+   selected resources as its contract.
+5. After every action or configuration change, run status again. The managed
+   loop performs this re-evaluation internally. A phase-local `next_action` can
+   continue that operation but never replaces the workspace Route.
+
+Explain, ask, confirm, and summarize in the user's current conversation
+language. Keep commands, flags, paths, ids, status values, JSONL keys,
+`source_ref` values, and copied CLI tokens unchanged.
+
+Do not infer repo sources, extraction scope, review decisions, or package output
+choices from surrounding files. Do not call source-repo operations such as
+clone, checkout, reset, fetch, install, build, or test without explicit user
+approval.

package/providers/context/actions/flush-runtime-events.yaml

@@ -0,0 +1,5 @@
+schema: agent-graph.action.v1
+id: flush-runtime-events
+runner: host
+effect: external
+handler: context.logs.flush

package/providers/context/codes.yaml

@@ -39,3 +39,4 @@ codes:
   - { code: route.package.configuration-required, kind: route-reason, summary: Declare the confirmed package output in the Context project., document: resources/procedures/package-output.md }
   - { code: route.package.template-review-required, kind: route-reason, summary: "Replace or edit package templates, or explicitly accept the generic defaults.", document: resources/procedures/package-output.md }
   - { code: route.build.package-stale, kind: route-reason, summary: Build the declared package outputs from current approved knowledge., document: resources/procedures/close-and-build.md }
+  - { code: route.logs.delivery-pending, kind: route-reason, summary: Send locally queued runtime logs through the configured generic sink., document: resources/procedures/close-and-build.md }