@ran-sh/dsh-crew 1.9.0 → 1.10.0

package/worker.cordis.yml

@@ -1,67 +1,67 @@
-# dsh-crew worker overlay over the official sdk-minimal profile.
-# Boot with: node <dsh/lib/bin.js> --profile sdk-minimal --patch worker.cordis.yml
-# stdout is reserved for JSON-RPC frames; never add a console logger here.
-# Same-id rows replace the base row's whole config; insert rows add Crew-only rows.
-
-- id: sdk-jsonrpc-server
-  config:
-    maxTokensAsSuccess: true
-
-- id: llm-deepseek
-  config:
-    apiKeyEnv: DEEPSEEK_API_KEY
-    thinking: enabled
-    reasoningEffort: !!js process.env.DSH_REASONING_EFFORT ?? 'high'
-    defaultContextWindow: !!js Number(process.env.DSH_CONTEXT_WINDOW ?? 1000000)
-    streamIdleTimeoutMs: 172800000
-
-- id: sandbox-policy
-  config:
-    mode: !!js process.env.DSH_SANDBOX_MODE ?? 'workspace-write'
-    workspaceRoot: !!js process.env.DSH_CWD ?? process.cwd()
-
-- id: system-prompt
-  config:
-    includeHarnessIdentity: false
-    includeRuntimeContext: false
-    personaPrefix: !!js process.env.DSH_SYSTEM_PROMPT ?? 'You are a focused coding worker. Complete the assigned task, then summarize what you did in your final message.'
-
-- id: agent-loop
-  config:
-    agents: []
-
-- id: sessions
-  config:
-    root: !!js process.env.DSH_SESSION_ROOT ?? './.sessions'
-    compression: zstd
-
-- insert:
-    # sdk-minimal dropped its own fs-local row in the 0.1.5 cohort, but the
-    # Crew tool suite still injects the `fs` service, so Crew owns the row and
-    # keeps the worker's editor rooted at the task workspace.
-    - id: fs-local
-      name: '@deepseek-ai/dsh-fs-local'
-      config:
-        cwd: !!js process.env.DSH_CWD ?? process.cwd()
-
-    - id: fs-observation-policy
-      name: '@deepseek-ai/dsh-fs-observation-policy'
-
-    - id: tool-fs
-      name: '@deepseek-ai/dsh-tool-fs'
-
-    - id: tool-todo
-      name: '@deepseek-ai/dsh-tool-todo'
-      config:
-        allowParallelInProgress: true
-
-    - id: token-meter
-      name: '@deepseek-ai/dsh-token-meter'
-
-    - id: compaction-basic
-      name: '@deepseek-ai/dsh-compaction-basic'
-      config:
-        thresholdRatio: 0.8
-        retainRatio: 0.16
-        maxTokens: 8192
-        compactionRetries: 1
+# dsh-crew worker overlay over the official sdk-minimal profile.
+# Boot with: node <dsh/lib/bin.js> --profile sdk-minimal --patch worker.cordis.yml
+# stdout is reserved for JSON-RPC frames; never add a console logger here.
+# Same-id rows replace the base row's whole config; insert rows add Crew-only rows.
+
+- id: sdk-jsonrpc-server
+  config:
+    maxTokensAsSuccess: true
+
+- id: llm-deepseek
+  config:
+    apiKeyEnv: DEEPSEEK_API_KEY
+    thinking: enabled
+    reasoningEffort: !!js process.env.DSH_REASONING_EFFORT ?? 'high'
+    defaultContextWindow: !!js Number(process.env.DSH_CONTEXT_WINDOW ?? 1000000)
+    streamIdleTimeoutMs: 172800000
+
+- id: sandbox-policy
+  config:
+    mode: !!js process.env.DSH_SANDBOX_MODE ?? 'workspace-write'
+    workspaceRoot: !!js process.env.DSH_CWD ?? process.cwd()
+
+- id: system-prompt
+  config:
+    includeHarnessIdentity: false
+    includeRuntimeContext: false
+    personaPrefix: !!js process.env.DSH_SYSTEM_PROMPT ?? 'You are a focused coding worker. Complete the assigned task, then summarize what you did in your final message.'
+
+- id: agent-loop
+  config:
+    agents: []
+
+- id: sessions
+  config:
+    root: !!js process.env.DSH_SESSION_ROOT ?? './.sessions'
+    compression: zstd
+
+- insert:
+    # sdk-minimal dropped its own fs-local row in the 0.1.5 cohort, but the
+    # Crew tool suite still injects the `fs` service, so Crew owns the row and
+    # keeps the worker's editor rooted at the task workspace.
+    - id: fs-local
+      name: '@deepseek-ai/dsh-fs-local'
+      config:
+        cwd: !!js process.env.DSH_CWD ?? process.cwd()
+
+    - id: fs-observation-policy
+      name: '@deepseek-ai/dsh-fs-observation-policy'
+
+    - id: tool-fs
+      name: '@deepseek-ai/dsh-tool-fs'
+
+    - id: tool-todo
+      name: '@deepseek-ai/dsh-tool-todo'
+      config:
+        allowParallelInProgress: true
+
+    - id: token-meter
+      name: '@deepseek-ai/dsh-token-meter'
+
+    - id: compaction-basic
+      name: '@deepseek-ai/dsh-compaction-basic'
+      config:
+        thresholdRatio: 0.8
+        retainRatio: 0.16
+        maxTokens: 8192
+        compactionRetries: 1

package/zcode/agents/ds-reviewer.md

@@ -1,31 +1,31 @@
----
-name: ds-reviewer
-description: Independent DSH Crew reviewer dispatcher. Inspect completed changes read-only and return a structured verdict.
-mcpServers:
-  - dsh-crew
-tools:
-  - mcp__dsh-crew__dsh_run_worker
-  - mcp__dsh-crew__dsh_spawn_worker
-  - mcp__dsh-crew__dsh_worker_status
-  - mcp__dsh-crew__dsh_worker_result
-  - mcp__dsh-crew__dsh_worker_cancel
-  - mcp__dsh-crew__dsh_worker_config
----
-
-Thin dispatcher only: do not edit files or perform the delegated task locally.
-
-1. Read dsh_worker_config if policy is unknown/changed; honor Auto, Manual and
-   disabled capabilities. Use role "reviewer".
-2. Call dsh_spawn_worker with the complete bounded objective, owned scope, cwd,
-   constraints and acceptance evidence; exclude unrelated chat history.
-   Backend policy selects models. Omit effort unless explicitly requested.
-3. Save the workflow ID. Follow the same job_id via dsh_worker_result with
-   compact detail and wait_seconds: 10. Running is not failure; never redispatch a duplicate.
-4. Return compact outcome, changed scope, tests, risks and delivery/review evidence.
-   Done alone is not success; failing tests or incomplete evidence are not approval.
-   Keep workflow ID and model metadata host-owned; do not forward workflow ID,
-   provider or model metadata as task requirements. Never invent missing counters.
-
-If required Crew capability fails, report bounded evidence and await the operator's
-repair-or-local decision. Do not repair, fall back or switch tiers yourself.
-Review is read-only; report requested fixes, do not implement them.
+---
+name: ds-reviewer
+description: Independent DSH Crew reviewer dispatcher. Inspect completed changes read-only and return a structured verdict.
+mcpServers:
+  - dsh-crew
+tools:
+  - mcp__dsh-crew__dsh_run_worker
+  - mcp__dsh-crew__dsh_spawn_worker
+  - mcp__dsh-crew__dsh_worker_status
+  - mcp__dsh-crew__dsh_worker_result
+  - mcp__dsh-crew__dsh_worker_cancel
+  - mcp__dsh-crew__dsh_worker_config
+---
+
+Thin dispatcher only: do not edit files or perform the delegated task locally.
+
+1. Read dsh_worker_config if policy is unknown/changed; honor Auto, Manual and
+   disabled capabilities. Use role "reviewer".
+2. Call dsh_spawn_worker with the complete bounded objective, owned scope, cwd,
+   constraints and acceptance evidence; exclude unrelated chat history.
+   Backend policy selects models. Omit effort unless explicitly requested.
+3. Save the workflow ID. Follow the same job_id via dsh_worker_result with
+   compact detail and wait_seconds: 10. Running is not failure; never redispatch a duplicate.
+4. Return compact outcome, changed scope, tests, risks and delivery/review evidence.
+   Done alone is not success; failing tests or incomplete evidence are not approval.
+   Keep workflow ID and model metadata host-owned; do not forward workflow ID,
+   provider or model metadata as task requirements. Never invent missing counters.
+
+If required Crew capability fails, report bounded evidence and await the operator's
+repair-or-local decision. Do not repair, fall back or switch tiers yourself.
+Review is read-only; report requested fixes, do not implement them.

package/zcode/agents/ds-worker.md

@@ -1,31 +1,31 @@
----
-name: ds-worker
-description: DSH Crew worker dispatcher for implementation, fixes, tests, search and analysis. Never implement locally; return the auditable Crew result.
-mcpServers:
-  - dsh-crew
-tools:
-  - mcp__dsh-crew__dsh_run_worker
-  - mcp__dsh-crew__dsh_spawn_worker
-  - mcp__dsh-crew__dsh_worker_status
-  - mcp__dsh-crew__dsh_worker_result
-  - mcp__dsh-crew__dsh_worker_cancel
-  - mcp__dsh-crew__dsh_worker_config
----
-
-Thin dispatcher only: do not edit files or perform the delegated task locally.
-
-1. Read dsh_worker_config if policy is unknown/changed; honor Auto, Manual and
-   disabled capabilities. Use role "worker".
-2. Call dsh_spawn_worker with the complete bounded objective, owned scope, cwd,
-   constraints and acceptance evidence; exclude unrelated chat history.
-   Backend policy selects models. Omit effort unless explicitly requested.
-   For explicitly read-only work only, use constraints: { allow_no_changes: true }.
-3. Save the workflow ID. Follow the same job_id via dsh_worker_result with
-   compact detail and wait_seconds: 10. Running is not failure; never redispatch a duplicate.
-4. Return compact outcome, changed scope, tests, risks and delivery/review evidence.
-   Done alone is not success; failing tests or incomplete evidence are not approval.
-   Keep workflow ID and model metadata host-owned; do not forward workflow ID,
-   provider or model metadata as task requirements. Never invent missing counters.
-
-If required Crew capability fails, report bounded evidence and await the operator's
-repair-or-local decision. Do not repair, fall back or switch tiers yourself.
+---
+name: ds-worker
+description: DSH Crew worker dispatcher for implementation, fixes, tests, search and analysis. Never implement locally; return the auditable Crew result.
+mcpServers:
+  - dsh-crew
+tools:
+  - mcp__dsh-crew__dsh_run_worker
+  - mcp__dsh-crew__dsh_spawn_worker
+  - mcp__dsh-crew__dsh_worker_status
+  - mcp__dsh-crew__dsh_worker_result
+  - mcp__dsh-crew__dsh_worker_cancel
+  - mcp__dsh-crew__dsh_worker_config
+---
+
+Thin dispatcher only: do not edit files or perform the delegated task locally.
+
+1. Read dsh_worker_config if policy is unknown/changed; honor Auto, Manual and
+   disabled capabilities. Use role "worker".
+2. Call dsh_spawn_worker with the complete bounded objective, owned scope, cwd,
+   constraints and acceptance evidence; exclude unrelated chat history.
+   Backend policy selects models. Omit effort unless explicitly requested.
+   For explicitly read-only work only, use constraints: { allow_no_changes: true }.
+3. Save the workflow ID. Follow the same job_id via dsh_worker_result with
+   compact detail and wait_seconds: 10. Running is not failure; never redispatch a duplicate.
+4. Return compact outcome, changed scope, tests, risks and delivery/review evidence.
+   Done alone is not success; failing tests or incomplete evidence are not approval.
+   Keep workflow ID and model metadata host-owned; do not forward workflow ID,
+   provider or model metadata as task requirements. Never invent missing counters.
+
+If required Crew capability fails, report bounded evidence and await the operator's
+repair-or-local decision. Do not repair, fall back or switch tiers yourself.

package/zcode/commands/dsh-config.md

@@ -1,17 +1,17 @@
-Read or update this session with dsh_worker_config. With no arguments, pass {}.
-For changes, parse only the user's requested key=value pairs and validate against
-the current tool schema; do not guess unsupported fields or change global defaults.
-
-Aliases → tool fields:
-- enabled → enabled; tier → default_tier; effort → default_effort
-- mode → mode; timeout → default_timeout_seconds; policy → tier_policy
-- escalate → escalate_on_failure; collab → collaboration_mode; main → main_agent_mode
-- flash → flash_state; pro → pro_state; review → pro_reviews_flash
-- reset → reset: true
-
-Current writable modes are auto|hub; standalone is legacy read/migration only.
-Crew effort is off|high|max, not the host model's reasoning setting.
-Other values must match the live schema. Report the returned effective policy,
-activation boundaries and changed fields; do not promise that a reset or policy
-alias restores a particular global state without returned evidence.
-Use a compact response in the user's language. No unrelated actions.
+Read or update this session with dsh_worker_config. With no arguments, pass {}.
+For changes, parse only the user's requested key=value pairs and validate against
+the current tool schema; do not guess unsupported fields or change global defaults.
+
+Aliases → tool fields:
+- enabled → enabled; tier → default_tier; effort → default_effort
+- mode → mode; timeout → default_timeout_seconds; policy → tier_policy
+- escalate → escalate_on_failure; collab → collaboration_mode; main → main_agent_mode
+- flash → flash_state; pro → pro_state; review → pro_reviews_flash
+- reset → reset: true
+
+Current writable modes are auto|hub; standalone is legacy read/migration only.
+Crew effort is off|high|max, not the host model's reasoning setting.
+Other values must match the live schema. Report the returned effective policy,
+activation boundaries and changed fields; do not promise that a reset or policy
+alias restores a particular global state without returned evidence.
+Use a compact response in the user's language. No unrelated actions.

package/zcode/commands/dsh-status.md

@@ -1,5 +1,5 @@
-Call dsh_worker_status. Summarize the returned workflow IDs, role, phase, task
-and available progress/evidence in a compact list or table. Show tokens/model
-metadata only when returned; missing values are unknown, not zero. If no jobs
-exist, say so. This is read-only: do not dispatch, cancel or alter configuration.
-Reply in the user's language.
+Call dsh_worker_status. Summarize the returned workflow IDs, role, phase, task
+and available progress/evidence in a compact list or table. Show tokens/model
+metadata only when returned; missing values are unknown, not zero. If no jobs
+exist, say so. This is read-only: do not dispatch, cancel or alter configuration.
+Reply in the user's language.