litclaude-ai 0.4.4 → 0.4.6

package/CHANGELOG.md

@@ -1,5 +1,27 @@
 # Changelog
 
+## 0.4.6 - 2026-09-03 — reader-facing communication boundary
+
+- Add request-scoped `reader`, `technical`, and `audit` communication modes across
+  Claude-native skills, commands, agents, hook context, child returns, and parent
+  synthesis. Reader mode is the default and keeps result, material risk, required
+  action, and explicitly requested detail while suppressing routine operational noise.
+- Preserve detailed execution-plane records, evidence, ledgers, checkpoints, review
+  packets, handoffs, and protected installer, doctor, status, debug, JSON, and audit
+  outputs. Add contract and real-surface coverage without weakening verification.
+- Keep enforcement honest: Claude Code exposes no supported final-response or
+  child-result interceptor, so generated prose follows the native prompt contract
+  advisory path rather than a generic scrubber or streamed-output rewrite.
+
+## 0.4.5 - 2026-09-02 — vendored-corpus companions and Skill-id namespace
+
+- Record Claude Code plugin-namespaced Skill tool ids (`plugin:skill`) as the
+  unprefixed skill id so skill-loop consultations match `SKILL_ID`.
+- Count vendor license, provenance, and NOTICE companions in family
+  payload-parity so packed skills keep the material they name after the
+  sibling vendor-layout moves.
+- This is a local release candidate only; no publish, tag, or push was performed.
+
 ## 0.4.4 - 2026-09-02 — inode-reuse hardening and Linux rotation safety
 
 - Bind regular-file reads and state transitions to ctime/birthtime-aware identities,

package/README.md

@@ -1,4 +1,4 @@
-<p align="center"><img src="https://cdn.jsdelivr.net/npm/litclaude-ai@0.4.4/cover.png" width="100%" alt="LitClaude — Claude Code-native workflow package" /></p>
+<p align="center"><img src="https://cdn.jsdelivr.net/npm/litclaude-ai@0.4.6/cover.png" width="100%" alt="LitClaude — Claude Code-native workflow package" /></p>
 
 ```text
 ██╗     ██╗████████╗
@@ -22,7 +22,7 @@
 </p>
 <p align="center">
   <img src="https://img.shields.io/badge/npm-litclaude--ai-cb3837" alt="npm: litclaude-ai" />
-  <img src="https://img.shields.io/badge/version-0.4.4-2ea44f" alt="version 0.4.4" />
+  <img src="https://img.shields.io/badge/version-0.4.6-2ea44f" alt="version 0.4.6" />
   <img src="https://img.shields.io/badge/Claude%20Code-plugin-blueviolet" alt="Claude Code plugin" />
   <img src="https://img.shields.io/badge/license-MIT-blue" alt="MIT license" />
 </p>
@@ -92,13 +92,13 @@ npx --yes litclaude-ai@latest install
 For a reproducible install, pin the current package version:
 
 ```bash
-npm view litclaude-ai@0.4.4 version
+npm view litclaude-ai@0.4.6 version
 ```
 
-If that lookup returns `0.4.4`, the exact install is available:
+If that lookup returns `0.4.6`, the exact install is available:
 
 ```bash
-npx --yes litclaude-ai@0.4.4 install
+npx --yes litclaude-ai@0.4.6 install
 ```
 
 Otherwise, wait for explicit human publication before using that pin. Check the

package/README_ko-KR.md

@@ -1,4 +1,4 @@
-<p align="center"><img src="https://cdn.jsdelivr.net/npm/litclaude-ai@0.4.4/cover.png" width="100%" alt="LitClaude — Claude Code-native workflow package" /></p>
+<p align="center"><img src="https://cdn.jsdelivr.net/npm/litclaude-ai@0.4.6/cover.png" width="100%" alt="LitClaude — Claude Code-native workflow package" /></p>
 
 ```text
 ██╗     ██╗████████╗
@@ -22,7 +22,7 @@
 </p>
 <p align="center">
   <img src="https://img.shields.io/badge/npm-litclaude--ai-cb3837" alt="npm: litclaude-ai" />
-  <img src="https://img.shields.io/badge/version-0.4.4-2ea44f" alt="version 0.4.4" />
+  <img src="https://img.shields.io/badge/version-0.4.6-2ea44f" alt="version 0.4.6" />
   <img src="https://img.shields.io/badge/Claude%20Code-plugin-blueviolet" alt="Claude Code plugin" />
   <img src="https://img.shields.io/badge/license-MIT-blue" alt="MIT license" />
 </p>
@@ -94,13 +94,13 @@ npx --yes litclaude-ai@latest install
 뒤 고정합니다.
 
 ```bash
-npm view litclaude-ai@0.4.4 version
+npm view litclaude-ai@0.4.6 version
 ```
 
-조회 결과가 `0.4.4`이면 exact install을 사용할 수 있습니다.
+조회 결과가 `0.4.6`이면 exact install을 사용할 수 있습니다.
 
 ```bash
-npx --yes litclaude-ai@0.4.4 install
+npx --yes litclaude-ai@0.4.6 install
 ```
 
 그렇지 않으면 명시적인 human publication을 기다립니다. Pin은 그 뒤에 사용합니다.

package/RELEASE_CHECKLIST.md

@@ -1,6 +1,6 @@
 # LitClaude Release Checklist
 
-Status: `litclaude-ai@0.4.4` is the current release candidate — exact canonical
+Status: `litclaude-ai@0.4.6` is the current release candidate — exact canonical
 frontend corpus plus Claude-native `autoresearch`, `autoconference`, and `wikify`
 workflow-family integration. It byte-pins the frontend library, legal companions,
 family source closures, and adapters through independent commitments and package
@@ -45,9 +45,9 @@ side-effect-free, the launcher starts only a separate Claude Code
 print/background worker, and the release preserves the Korean polishing
 command, strict multi-agent review pipeline, fidelity guardrails, package
 hygiene checks, native route gates, and safe start-work handoff behavior.
-`package.json` is aligned to `0.4.4`,
-`plugins/litclaude/.claude-plugin/plugin.json` is aligned to `0.4.4`, and the
-plugin-local MCP server reports `0.4.4`.
+`package.json` is aligned to `0.4.6`,
+`plugins/litclaude/.claude-plugin/plugin.json` is aligned to `0.4.6`, and the
+plugin-local MCP server reports `0.4.6`.
 
 This candidate adds an approval-gated skill learning loop. Validated
 consultations, bounded corrections, iteration signals, and coverage gaps may
@@ -295,9 +295,9 @@ checkout and from an isolated install of the packed tarball:
 Before requesting publication approval, confirm these artifacts from the current
 checkout:
 
-- `package.json` version is `0.4.4`.
-- `plugins/litclaude/.claude-plugin/plugin.json` version is `0.4.4`.
-- `plugins/litclaude/bin/litclaude-mcp.js` reports server version `0.4.4`.
+- `package.json` version is `0.4.6`.
+- `plugins/litclaude/.claude-plugin/plugin.json` version is `0.4.6`.
+- `plugins/litclaude/bin/litclaude-mcp.js` reports server version `0.4.6`.
 - Prompt-hook tests cover bundled `SKILL.md` body injection for bare `hyperplan`, `litresearch`, `lit research`, `init-deep`, and explicit leading `$start-work`; diagnostic/copy mentions stay inert while leading natural-language `lit start work` stays BLOCKED.
 - `lit search` and `lit query` route to `/litclaude:litresearch` without activating on slash mentions, code spans, or non-lit prompts.
 - Litresearch web lanes require public API/feed preference, validator-first checks, route traces, prompt-injection quarantine, and honest auth/paywall/private-data stop reasons.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "litclaude-ai",
-  "version": "0.4.4",
+  "version": "0.4.6",
   "description": "Claude Code-native workflow distribution.",
   "type": "module",
   "bin": {

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

@@ -1,7 +1,7 @@
 {
   "name": "litclaude",
   "description": "Claude Code-native workflow plugin with a bounded-authority start-work lifecycle.",
-  "version": "0.4.4",
+  "version": "0.4.6",
   "author": {
     "name": "LitClaude contributors"
   },

package/plugins/litclaude/agents/boulder-executor.md

@@ -24,7 +24,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -51,10 +51,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.
@@ -77,7 +90,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 Execute only the assigned approved plan checkbox. Do not redesign, rescope, or
 replace the approved plan unless evidence proves a blocker or stale state. Write
 the failing test first, make the smallest change, run verification, capture
-real-surface evidence, and report cleanup. Your DoneClaim must name changed
+real-surface evidence, and report cleanup. Your internal DoneClaim must name changed
 files, exact tests, real-surface evidence, risks, and cleanup receipt.
 
 The code-owned start-work lifecycle is the executor's authority boundary. Read

package/plugins/litclaude/agents/korean-prose-editor.md

@@ -23,7 +23,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -50,10 +50,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/korean-style-analyzer.md

@@ -23,7 +23,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -50,10 +50,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/librarian-researcher.md

@@ -23,7 +23,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -50,10 +50,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/meaning-preservation-auditor.md

@@ -24,7 +24,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -51,10 +51,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/native-flow-reviewer.md

@@ -24,7 +24,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -51,10 +51,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/oracle-verifier.md

@@ -25,7 +25,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -52,10 +52,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/polish-orchestrator.md

@@ -24,7 +24,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -51,10 +51,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/prometheus-planner.md

@@ -24,7 +24,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -51,10 +51,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/qa-runner.md

@@ -24,7 +24,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -51,10 +51,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.

package/plugins/litclaude/agents/quality-reviewer.md

@@ -25,7 +25,7 @@ verdicts: [PASS, FAIL, BLOCKED]
 | --- | --- | --- |
 | activation | Accept only a bounded Claude Code subagent assignment. | Restate `TASK:`, `DELIVERABLE`, `SCOPE`, and `VERIFY`. |
 | tools | Use only tools granted by frontmatter and the active permission mode. | Mention skipped actions when permissions are read-only. |
-| completion | Return one final packet with status and proof. | Include `PASS`, `FAIL`, or `BLOCKED:` plus paths or commands. |
+| completion | Retain one final packet with status and proof. | Return its request-scoped projection; include paths or commands only when authorized. |
 
 ## #contract.inputs
 
@@ -52,10 +52,23 @@ verdicts: [PASS, FAIL, BLOCKED]
 
 ## #contract.outputs
 
+The bullets below are fields of the internal agent packet consumed by the parent,
+not a reader-facing template. Classify them as RESULT, RISK, ACTION,
+REQUESTED_DETAIL, or INTERNAL_METADATA before return; exact verification,
+evidence paths, and cleanup receipts are INTERNAL_METADATA unless the assigned
+return mode requests them or they carry a material risk or required action.
+
 - `TASK:` summary, delivered artifact or findings, and exact `VERIFY` result.
-- Evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
+- Detailed internal evidence list with file paths, command transcripts, hook/package/MCP/LSP probes, or Manual-QA artifacts.
 - Cleanup receipt when any process, session, temp path, browser, port, or worktree was used.
 
+Return-mode contract: only an explicit parent-to-child return mode may select
+technical or audit detail; the default is reader. A reader return contains
+conclusions, material risks, unresolved issues, required actions, and requested
+artifacts. Search logs, command diaries, evidence paths, timestamps, and reasoning
+chronology remain internal unless requested. The child cannot elevate the parent
+mode; parent synthesis filters the packet again.
+
 ## #contract.evidence
 
 - Prefer fresh local reads and replayable commands over stale handoff claims.