soturail 1.2.0 → 1.4.0

package/docs/rails/skills/skill-rail-2.md

@@ -0,0 +1,42 @@
+# Skill Rail 2.0
+
+Skill Rail 2.0 builds reviewed, source-mapped local skill packs. Skills are operating procedures for agent hosts, not self-modifying behavior.
+
+## Commands
+
+```bash
+soturail skills template docs-review
+soturail skills build README.md docs --name project-guide
+soturail skills fold-in project-guide docs/new-guide.md
+soturail skills lint
+soturail skills eval
+soturail skills report
+```
+
+The existing `skills init|list|validate|suggest|route|export|pack` commands remain available.
+
+## Pack Layout
+
+```txt
+.soturail/skills/<skill-id>/
+  SKILL.md
+  skill.yml
+  safety.md
+  examples/
+  topics/
+  glossary.md
+  patterns.md
+  cheatsheet.md
+  metadata.json
+  source-map.json
+```
+
+Templates include supported hosts, risk level, examples, verification steps and safety boundaries. Built skills use Knowledge Rail output so generated topics and source maps remain locally traceable.
+
+## Validation
+
+Lint and evaluation check metadata, supported hosts, risk, examples, verification steps, source maps, unsafe commands, secret access, remote writes and MCP exposure warnings.
+
+No skill should hide destructive commands or imply that SotuRail is an autonomous agent runtime.
+
+Related: [Knowledge Rail](../knowledge/knowledge-rail.md), [Tasklet Rail](../tasklets/tasklet-rail.md), [Policy Rail](../governance/policy-rail.md).

package/docs/rails/tasklets/tasklet-rail.md

@@ -0,0 +1,32 @@
+# Tasklet Rail
+
+Tasklet Rail manages small reusable local task templates. A tasklet describes bounded work; it is not an autonomous agent or task runner.
+
+## Commands
+
+```bash
+soturail tasklet create review-docs
+soturail tasklet list
+soturail tasklet run review-docs --dry-run
+soturail tasklet export review-docs
+```
+
+Tasklets live under:
+
+```txt
+.soturail/tasklets/<name>.md
+```
+
+Each template includes:
+
+- Objective
+- Allowed context
+- Allowed files
+- Disallowed actions
+- Verification commands
+- Definition of done
+- Expected handoff
+
+`tasklet run` always returns a dry-run simulation and never executes shell commands. `tasklet export` writes an agent-readable handoff under `.soturail/exports/tasklets/`.
+
+Related: [Skill Rail 2.0](../skills/skill-rail-2.md), [Workflow Rail](../harness/workflow-rail.md), [Security Boundaries](../../security/security-boundaries.md).

package/docs/{release-checklist.md → reference/commands/release-checklist.md}

@@ -96,7 +96,7 @@ docs/releases/RELEASE_NOTES_vX.Y.Z.md
 
 Do not paste Markdown prose or code-fence labels directly into `cmd.exe` as commands. For example, do not paste ```` ```bat ````. Copy only the command lines themselves.
 
-See [docs/windows.md](windows.md) for CMD and PowerShell quoting notes.
+See [docs/getting-started/windows.md](../../getting-started/windows.md) for CMD and PowerShell quoting notes.
 
 ## Automation
 

package/docs/{stable-command-surface.md → reference/commands/stable-command-surface.md}

@@ -1,6 +1,6 @@
 # Stable Command Surface
 
-SotuRail v1.0.0 froze the first stable local Context OS surface. SotuRail v1.1.0 kept that surface stable and added Host Compatibility Rail commands. SotuRail v1.2.0 adds the compatible local Harness Lifecycle command family.
+SotuRail v1.0.0 froze the first stable local Context OS surface. SotuRail v1.1.0 added Host Compatibility Rail, v1.2.0 added Harness Lifecycle Rail and v1.4.0 adds compatible Knowledge, Evidence, Agent QA, Skill Rail 2.0 and Tasklet command families.
 
 ## Stable Surface For v1.0
 
@@ -33,6 +33,12 @@ SotuRail v1.0.0 froze the first stable local Context OS surface. SotuRail v1.1.0
 - `soturail mcp exposure`
 - `soturail mcp resources report`
 - `soturail mcp resources host-manifest`
+- `soturail knowledge estimate|compile|update|verify|list`
+- `soturail evidence collect|verify|report`
+- `soturail eval dataset init|run`
+- `soturail eval golden|regression`
+- `soturail skills template|lint|eval|report|build|fold-in`
+- `soturail tasklet create|list|run --dry-run|export`
 
 Stable means the command should remain local, documented, covered by smoke or contract tests, and changed compatibly where possible.
 
@@ -45,7 +51,7 @@ Stable means the command should remain local, documented, covered by smoke or co
 - Large parser integrations.
 - Full Knowledge Graph runtime.
 - Full Design Rail runtime.
-- Skill Rail 2.0 runtime.
+- Autonomous skill or tasklet execution.
 - Governance and cost runtime.
 - SotuRail Conductor commands and autonomous/runtime behavior.
 
@@ -65,17 +71,18 @@ Advanced commands remain useful for diagnostics, but v1.0 compatibility focuses
 
 No public commands are deprecated in v1.0.0.
 
-Future deprecations must name a replacement command, state the release that introduced the warning and include removal timing in release notes. See [deprecation policy](deprecation-policy.md).
+Future deprecations must name a replacement command, state the release that introduced the warning and include removal timing in release notes. See [deprecation policy](../contracts/deprecation-policy.md).
 
 ## Related Contracts
 
-- [v1 contract](v1-contract.md)
-- [schema contracts](schema-contracts.md)
-- [migration to v1](migration-v1.md)
-- [agent hosts](agent-hosts.md)
-- [host matrix schema](host-matrix-schema.md)
-- [agent export contract](agent-export-contract.md)
-- [MCP host manifest](mcp-host-manifest.md)
-- [Harness Lifecycle Rail](harness-lifecycle-rail.md)
-- [Security Boundaries](security-boundaries.md)
-- [Conductor Mode](conductor-mode.md)
+- [v1 contract](../contracts/v1-contract.md)
+- [schema contracts](../schemas/schema-contracts.md)
+- [migration to v1](../../getting-started/migration-v1.md)
+- [agent hosts](../../rails/hosts/agent-hosts.md)
+- [host matrix schema](../schemas/host-matrix-schema.md)
+- [agent export contract](../contracts/agent-export-contract.md)
+- [MCP host manifest](../../rails/hosts/mcp-host-manifest.md)
+- [Harness Lifecycle Rail](../../rails/harness/harness-lifecycle-rail.md)
+- [Security Boundaries](../../security/security-boundaries.md)
+- [Conductor Mode](../../ecosystem/conductor-mode.md)
+- [v1.4 commands](v1.4-commands.md)

package/docs/reference/commands/v1.4-commands.md

@@ -0,0 +1,51 @@
+# v1.4 Command Reference
+
+## Knowledge
+
+```bash
+soturail knowledge estimate <paths...>
+soturail knowledge compile <paths...> --name <name>
+soturail knowledge update <name> <paths...>
+soturail knowledge verify <name>
+soturail knowledge list
+```
+
+## Evidence
+
+```bash
+soturail evidence collect
+soturail evidence verify
+soturail evidence report
+```
+
+## Agent QA
+
+```bash
+soturail eval dataset init
+soturail eval dataset run
+soturail eval golden
+soturail eval regression
+soturail eval report
+```
+
+## Skill Rail 2.0
+
+```bash
+soturail skills template <domain>
+soturail skills build <paths...> --name <name>
+soturail skills fold-in <skill> <paths...>
+soturail skills lint
+soturail skills eval
+soturail skills report
+```
+
+## Tasklets
+
+```bash
+soturail tasklet create <name>
+soturail tasklet list
+soturail tasklet run <name> --dry-run
+soturail tasklet export <name>
+```
+
+All commands are local-first. Knowledge and evaluation are deterministic; evidence does not invent verification; tasklet execution is simulation-only.

package/docs/{deprecation-policy.md → reference/contracts/deprecation-policy.md}

@@ -13,7 +13,7 @@ SotuRail v1.0.0 treats command and schema stability as a first-class release con
 
 ## Stable Commands
 
-Commands listed in [stable-command-surface.md](stable-command-surface.md) are the v1.0 stable surface. Compatible additions are allowed. Removals, meaning changes or output-shape breaks require a warning period and migration notes.
+Commands listed in [stable-command-surface.md](../commands/stable-command-surface.md) are the v1.0 stable surface. Compatible additions are allowed. Removals, meaning changes or output-shape breaks require a warning period and migration notes.
 
 ## Experimental Commands
 

package/docs/{v1-contract.md → reference/contracts/v1-contract.md}

@@ -1,6 +1,6 @@
 # v1 Contract
 
-SotuRail v1.0.0 froze the first stable local Context OS surface. SotuRail v1.1.0 extends that surface with host-compatible exports and read-only host manifests. SotuRail v1.2.0 adds compatible local harness lifecycle state while keeping experimental host runtimes and Conductor outside the stable contract.
+SotuRail v1.0.0 froze the first stable local Context OS surface. SotuRail v1.1.0 extends that surface with host-compatible exports and read-only host manifests. SotuRail v1.2.0 adds compatible local harness lifecycle state. SotuRail v1.4.0 adds deterministic Knowledge, Evidence, Agent QA, Skill Rail 2.0 and dry-run Tasklet contracts while keeping experimental host runtimes and Conductor outside the stable contract.
 
 ## Stable Promise
 
@@ -16,7 +16,7 @@ SotuRail v1.0.0 froze the first stable local Context OS surface. SotuRail v1.1.0
 
 ## Stable Commands
 
-The stable command surface is listed in [stable-command-surface.md](stable-command-surface.md). It includes status, reports, dashboard, observability, Project Brain, evaluation, benchmarks, native candidate reporting, baseline snapshots, release checks, workflow, harness lifecycle, diagrams, agent exports, host doctors and read-only MCP report/host resources.
+The stable command surface is listed in [stable-command-surface.md](../commands/stable-command-surface.md). It includes status, reports, dashboard, observability, Project Brain, evaluation, benchmarks, native candidate reporting, baseline snapshots, release checks, workflow, harness lifecycle, diagrams, agent exports, host doctors and read-only MCP report/host resources.
 
 ## Non-Goals
 
@@ -24,4 +24,4 @@ SotuRail does not add a cloud dashboard, hosted analytics, telemetry upload, log
 
 ## Compatibility
 
-Compatible changes may add fields to JSON artifacts. Removing fields, changing schema meanings or promoting experimental commands requires release notes and migration guidance. See [deprecation policy](deprecation-policy.md) and [migration to v1](migration-v1.md).
+Compatible changes may add fields to JSON artifacts. Removing fields, changing schema meanings or promoting experimental commands requires release notes and migration guidance. See [deprecation policy](deprecation-policy.md) and [migration to v1](../../getting-started/migration-v1.md).

package/docs/{licensing-strategy.md → reference/licensing-strategy.md}

@@ -74,7 +74,7 @@ The root `LICENSE` file should contain the full Apache License 2.0 text.
 Recommended README badge:
 
 ```md
-[![License: Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-blue)](LICENSE)
+[![License: Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-blue)](../../LICENSE)
 ```
 
 Recommended README language:

package/docs/{schema-contracts.md → reference/schemas/schema-contracts.md}

@@ -24,6 +24,18 @@ Stable v1 JSON artifacts are local files designed for humans, CI and coding agen
 - `soturail.harness.audit.v1`
 - `soturail.feature-list.v1`
 - `soturail.session.v1`
+- `soturail.knowledge.estimate.v1`
+- `soturail.knowledge.v1`
+- `soturail.knowledge.source-map.v1`
+- `soturail.knowledge.verify.v1`
+- `soturail.evidence.provenance.v1`
+- `soturail.eval.dataset.v1`
+- `soturail.agent-qa.v1`
+- `soturail.skills.report.v2`
+- `soturail.skill-pack.v2`
+- `soturail.skill-source-map.v1`
+- `soturail.tasklet.v1`
+- `soturail.tasklet.simulation.v1`
 
 ## Required Principles
 
@@ -53,6 +65,8 @@ soturail agents doctor --all --json
 soturail mcp resources host-manifest --host codex --json
 ```
 
-They must not grant mutation resources, shell execution through MCP, cloud telemetry or raw evidence exposure. See [host-matrix-schema.md](host-matrix-schema.md), [agent-export-contract.md](agent-export-contract.md) and [mcp-host-manifest.md](mcp-host-manifest.md).
+They must not grant mutation resources, shell execution through MCP, cloud telemetry or raw evidence exposure. See [host-matrix-schema.md](host-matrix-schema.md), [agent-export-contract.md](../contracts/agent-export-contract.md) and [mcp-host-manifest.md](../../rails/hosts/mcp-host-manifest.md).
 
-Harness Lifecycle JSON artifacts are optional until `soturail harness init` or `soturail session start` is run. Once present, schema checks validate their shape. See [Harness Lifecycle Rail](harness-lifecycle-rail.md).
+Harness Lifecycle JSON artifacts are optional until `soturail harness init` or `soturail session start` is run. Once present, schema checks validate their shape. See [Harness Lifecycle Rail](../../rails/harness/harness-lifecycle-rail.md).
+
+v1.4 knowledge, evidence, Agent QA, Skill Rail 2.0 and tasklet artifacts are generated only when their local commands run. They remain source-backed, parseable and safe for offline use. See [v1.4 commands](../commands/v1.4-commands.md).

package/docs/releases/RELEASE_NOTES_v0.10.1.md

@@ -51,9 +51,9 @@ soturail release check
 .soturail/schemas/check.md
 .soturail/readiness/v1.json
 .soturail/readiness/v1.md
-docs/stable-command-surface.md
-docs/deprecation-policy.md
-docs/migration-v1.md
+docs/reference/commands/stable-command-surface.md
+docs/reference/contracts/deprecation-policy.md
+docs/getting-started/migration-v1.md
 ```
 
 ## Stability And Safety

package/docs/releases/RELEASE_NOTES_v1.0.1.md

@@ -4,7 +4,7 @@
 
 - Updated package and repository licensing metadata to Apache-2.0.
 - Aligned README license badge and license section with Apache-2.0.
-- Aligned `docs/licensing-strategy.md` with the Apache-2.0 licensing decision.
+- Aligned `docs/reference/licensing-strategy.md` with the Apache-2.0 licensing decision.
 
 ## Notes
 

package/docs/releases/RELEASE_NOTES_v1.4.0.md

@@ -0,0 +1,47 @@
+# SotuRail v1.4.0
+
+SotuRail v1.4.0 combines Knowledge, Evidence, Evaluation, Skills and Tasklets into one source-backed local workflow. The previously planned v1.3.0 scope is absorbed into this release because useful skills depend on knowledge compilation, provenance and deterministic evaluation.
+
+## Added
+
+- `soturail knowledge estimate|compile|update|verify|list`
+- `soturail evidence collect|verify|report`
+- `soturail eval dataset init|run`
+- `soturail eval golden|regression`
+- `soturail skills template|lint|eval|report|build|fold-in`
+- `soturail tasklet create|list|run --dry-run|export`
+- Local knowledge packs, provenance reports, Agent QA fixtures, Skill Rail 2.0 packs and tasklet exports
+- Documentation index and Markdown link checker
+
+## Changed
+
+- Reorganized documentation into getting-started, rail, reference, architecture, security, ecosystem, roadmap, release and tutorial sections.
+- Rewrote README as a concise landing page.
+- Updated roadmap direction so v1.3.0 is explicitly absorbed into v1.4.0.
+
+## Safety
+
+- No embeddings, cloud calls, required model provider or mandatory database.
+- No unsupported verification claims.
+- No autonomous tasklet execution or hidden shell commands.
+- No required server or destructive MCP expansion.
+
+## Verification
+
+```bash
+npm install
+npm run typecheck
+npm run build
+npm test
+npm run docs:check
+npm audit --audit-level=high
+npm pack --dry-run
+npm run release:check
+```
+
+## Known Limitations
+
+- Knowledge summaries are deterministic local extracts, not semantic model summaries.
+- Evidence verification checks recorded local artifacts; it does not execute verification commands.
+- LLM-as-judge remains optional and documentation-only.
+- Knowledge Graph, Host Router, governance/cost and Conductor runtimes remain planned.

package/docs/{future-rails-index.md → roadmap/future-rails-index.md}

@@ -24,8 +24,8 @@ v1.0.0 froze the first stable local Context OS surface. v1.1.0 delivered Host Co
 ```txt
 v1.1.1  Host Compatibility Polish, ecosystem docs and golden export checks
 v1.2.0  Spec, Design, Diagram and Harness Lifecycle Rail
-v1.3.0  Knowledge, Evidence and Evaluation Rail
-v1.4.0  Skill Rail 2.0, Knowledge-to-Skill and Tasklet Packs
+v1.3.0  Absorbed into v1.4.0
+v1.4.0  Knowledge, Evidence, Evaluation, Skills and Tasklets
 v1.5.0  Governance, Cost, Resilience and Host Router Rail
 v1.6.0  Agent Governance / Evolution Rail
 ```
@@ -34,7 +34,7 @@ v1.2.0 also delivers a focused Harness Lifecycle slice: safe scaffold initializa
 
 ## 2026 Agent Runtime Update
 
-The newest planning layer is tracked in [`docs/roadmap-agent-runtime-addendum.md`](roadmap-agent-runtime-addendum.md).
+The newest planning layer is tracked in [`docs/roadmap/roadmap-agent-runtime-addendum.md`](roadmap-agent-runtime-addendum.md).
 
 It tightens how the future rails connect around current agent-runtime patterns:
 
@@ -50,42 +50,42 @@ It tightens how the future rails connect around current agent-runtime patterns:
 
 | Rail | Purpose | Main version target | Primary docs |
 | --- | --- | --- | --- |
-| Memory Rail | Store, recall, capture and consolidate approved local memory | v0.5.0+ | `docs/memory-rail.md`, `docs/context-packs.md`, `ROADMAP.md` |
-| Context Intelligence | Select, prune, route and offload task-specific context | v0.5.0+ | `docs/context-packs.md`, `docs/ecosystem-influences.md` |
-| Context Expert Router | Route tasks to code/docs/release/security/workflow/memory/research context experts | v0.5.0+ | `docs/context-packs.md`, `ROADMAP.md` |
-| Context Budget Rail | Estimate context drivers and recommend compaction/offload before agent handoff | v0.5.0 seed | `docs/roadmap-agent-runtime-addendum.md`, `docs/context-packs.md` |
-| Role Packs | Generate planner/executor/reviewer/release-manager/researcher packs | v0.5.0+ | `docs/context-packs.md`, `docs/workflow-rail.md` |
-| Run Workspace Rail | Store per-run input, output, raw logs, offloads, artifacts, summaries and evidence | v0.5.0 seed | `docs/roadmap-agent-runtime-addendum.md`, `docs/workflow-rail.md` |
-| MCP Exposure Rail | Report exposed MCP tools/resources/prompts/roots and local risk notes | v0.5.0 seed | `docs/roadmap-agent-runtime-addendum.md`, `docs/mcp.md`, `docs/policy-rail.md` |
-| Skill Boundary Rail | Route skills by task, role, evidence and host capability instead of always loading them | v0.5.0 seed | `docs/roadmap-agent-runtime-addendum.md`, `docs/skill-rail.md` |
-| Harness Rail | setup/plan/work/review/release discipline, evidence packs and failure ledger | v0.5.0 seeds, v0.7.0 expansion | `docs/harness-rail.md`, `docs/workflow-rail.md` |
-| Harness Lifecycle Rail | Safe local instructions, state, sessions, features, audits and handoffs | v1.2.0 implemented | `docs/harness-lifecycle-rail.md`, `docs/security-boundaries.md` |
-| Acceptance Harness Contracts | Require build/typecheck/lint/test/coverage/docs/policy gates before accepting work | v0.5.0 seed, v0.7.0 expansion | `docs/roadmap-agent-runtime-addendum.md`, `docs/harness-rail.md` |
-| Policy Rail | Rules, approvals, risky action queue, auth checks and MCP exposure reports | v0.5.0+ | `docs/policy-rail.md`, `docs/security-model.md`, `docs/rules.md` |
-| Structured Payload Rail | Target-aware Markdown/JSON/tagged/TOON-table/Mermaid context formats | v0.5.1+ | `docs/structured-payload-rail.md`, `docs/context-packs.md` |
-| Diagram Rail | Mermaid diagrams and `.spec.md` visual contracts | v0.5.1 docs, v0.7.0 expansion | `docs/diagram-rail.md`, `docs/spec-driven-workflow.md` |
-| Agent Docs Hygiene | Keep `CLAUDE.md`, `AGENTS.md`, `GEMINI.md` and Cursor rules short and referenced | v0.5.0+ | `docs/agent-docs-hygiene.md`, `docs/agents.md` |
-| Filesystem Evidence Rail | Snapshot, touched-file and diff evidence tied to workflows and raw IDs | v0.5.0 seeds, v0.7.0 expansion | `docs/filesystem-evidence-rail.md`, `docs/workflow-rail.md` |
-| Evidence Pack | Connect build/test/audit/pack/raw IDs/policy decisions/release notes | v0.5.0 seeds, v0.7.0 expansion | `docs/harness-rail.md`, `docs/workflow-rail.md` |
-| Evaluation Suite | Quality benchmarks for context, payload formats, diagrams, evidence and role packs | v0.6.1 | `docs/evaluation-suite.md`, `docs/benchmarking.md`, `ROADMAP.md` |
-| Reverse Specification Rail | Turn existing code/docs/config/logs into claims, rules, specs, gaps and validation tasks | v0.8.0 primary | `docs/reverse-specification-rail.md`, `docs/knowledge-to-rules.md` |
-| Project Brain | Verified claims, decisions, bugs, gaps, rules, stale events and agent-safe briefs | v0.8.0 | `docs/project-brain.md`, `ROADMAP.md` |
-| Host Compatibility Rail | Host-aware exports, capability matrix, host doctors and conservative prompt-only fallback for OpenCode, Antigravity, Claude, Codex, Cursor, Deep Agents-style and generic hosts | v1.1.0 implemented | `docs/host-compatibility-rail.md`, `docs/agents.md`, `docs/mcp.md`, `docs/agent-export-contract.md` |
-| Agent QA Rail | Offline datasets, golden exports, regression reports and optional LLM-as-judge policy | v1.1.1 docs, v1.3.0 planned | `docs/agent-qa-rail.md`, `docs/eval-datasets.md`, `docs/golden-agent-tests.md`, `docs/llm-as-judge-policy.md` |
-| Evidence And Provenance Rail | Provenance sidecars, verification statuses and evidence per run/report | v1.3.0 | `docs/evidence-provenance-rail.md`, `docs/report-rail.md`, `docs/filesystem-evidence-rail.md` |
-| Knowledge Rail | Compile docs/specs/notes into on-demand SKILL.md/topic/glossary/pattern packs | v1.3.0 docs, v1.4.0 skill integration | `docs/knowledge-rail.md`, `docs/skill-rail-2.md`, `docs/project-brain.md` |
-| Resilience Rail | Rate-limit, fallback and provider-risk documentation/reporting without proxying traffic | v1.5.0 | `docs/resilience-rail.md`, `docs/rate-limit-and-fallback-policy.md`, `docs/governance-cost-rail.md` |
-| Host Router Rail | Route one local context source into host-specific export formats with safe fallback | v1.5.0 | `docs/host-router-rail.md`, `docs/host-compatibility-rail.md`, `docs/context-packs.md` |
-| Tasklet Rail | Small local reusable task templates for agents | v1.4.0 exploration | `docs/tasklet-rail.md`, `docs/skill-rail-2.md`, `docs/workflow-rail.md` |
-| Agent Governance/Evolution Rail | Trace, ledger, experiments, approval gates and improve/eval/apply loop | v1.6.0 | `docs/agent-governance-rail.md`, `docs/evaluation-suite.md`, `docs/policy-rail.md` |
-| Spec Rail | PRD, requirements, design, tasks and acceptance criteria as workflow inputs | v1.2.0 | `docs/spec-driven-workflow.md`, `docs/design-rail.md`, `docs/diagram-rail.md` |
-| Design Rail | Local `DESIGN.md`, design token lint/diff/export and agent-readable visual guidance | v1.2.0 | `docs/design-rail.md`, `docs/dashboard-rail.md` |
-| Knowledge Graph Rail | Local graph of files, claims, decisions, tests, workflows, diagrams and releases | v1.3.0 | `docs/knowledge-graph-rail.md`, `docs/code-graph.md`, `docs/project-brain.md` |
-| Skill Rail 2.0 | Domain skill templates, lint/eval/report and role-aware exports | v1.4.0 | `docs/skill-rail-2.md`, `docs/skill-rail.md`, `docs/policy-rail.md` |
-| Governance And Cost Rail | Context budget, dynamic workflow risk and MCP/skill exposure warnings | v1.5.0 | `docs/governance-cost-rail.md`, `docs/report-rail.md`, `docs/policy-rail.md` |
-| Redacted Evidence Sharing | Local redacted evidence bundles for reports/logs without hosting by default | v1.4+ exploration | `docs/report-redaction.md`, `docs/external-projects-audit.md` |
+| Memory Rail | Store, recall, capture and consolidate approved local memory | v0.5.0+ | `docs/rails/context/memory-rail.md`, `docs/rails/context/context-packs.md`, `ROADMAP.md` |
+| Context Intelligence | Select, prune, route and offload task-specific context | v0.5.0+ | `docs/rails/context/context-packs.md`, `docs/ecosystem/ecosystem-influences.md` |
+| Context Expert Router | Route tasks to code/docs/release/security/workflow/memory/research context experts | v0.5.0+ | `docs/rails/context/context-packs.md`, `ROADMAP.md` |
+| Context Budget Rail | Estimate context drivers and recommend compaction/offload before agent handoff | v0.5.0 seed | `docs/roadmap/roadmap-agent-runtime-addendum.md`, `docs/rails/context/context-packs.md` |
+| Role Packs | Generate planner/executor/reviewer/release-manager/researcher packs | v0.5.0+ | `docs/rails/context/context-packs.md`, `docs/rails/harness/workflow-rail.md` |
+| Run Workspace Rail | Store per-run input, output, raw logs, offloads, artifacts, summaries and evidence | v0.5.0 seed | `docs/roadmap/roadmap-agent-runtime-addendum.md`, `docs/rails/harness/workflow-rail.md` |
+| MCP Exposure Rail | Report exposed MCP tools/resources/prompts/roots and local risk notes | v0.5.0 seed | `docs/roadmap/roadmap-agent-runtime-addendum.md`, `docs/rails/hosts/mcp.md`, `docs/rails/governance/policy-rail.md` |
+| Skill Boundary Rail | Route skills by task, role, evidence and host capability instead of always loading them | v0.5.0 seed | `docs/roadmap/roadmap-agent-runtime-addendum.md`, `docs/rails/skills/skill-rail.md` |
+| Harness Rail | setup/plan/work/review/release discipline, evidence packs and failure ledger | v0.5.0 seeds, v0.7.0 expansion | `docs/rails/harness/harness-rail.md`, `docs/rails/harness/workflow-rail.md` |
+| Harness Lifecycle Rail | Safe local instructions, state, sessions, features, audits and handoffs | v1.2.0 implemented | `docs/rails/harness/harness-lifecycle-rail.md`, `docs/security/security-boundaries.md` |
+| Acceptance Harness Contracts | Require build/typecheck/lint/test/coverage/docs/policy gates before accepting work | v0.5.0 seed, v0.7.0 expansion | `docs/roadmap/roadmap-agent-runtime-addendum.md`, `docs/rails/harness/harness-rail.md` |
+| Policy Rail | Rules, approvals, risky action queue, auth checks and MCP exposure reports | v0.5.0+ | `docs/rails/governance/policy-rail.md`, `docs/security/security-model.md`, `docs/rails/governance/rules.md` |
+| Structured Payload Rail | Target-aware Markdown/JSON/tagged/TOON-table/Mermaid context formats | v0.5.1+ | `docs/rails/context/structured-payload-rail.md`, `docs/rails/context/context-packs.md` |
+| Diagram Rail | Mermaid diagrams and `.spec.md` visual contracts | v0.5.1 docs, v0.7.0 expansion | `docs/rails/design/diagram-rail.md`, `docs/rails/design/spec-driven-workflow.md` |
+| Agent Docs Hygiene | Keep `CLAUDE.md`, `AGENTS.md`, `GEMINI.md` and Cursor rules short and referenced | v0.5.0+ | `docs/rails/hosts/agent-docs-hygiene.md`, `docs/rails/hosts/agents.md` |
+| Filesystem Evidence Rail | Snapshot, touched-file and diff evidence tied to workflows and raw IDs | v0.5.0 seeds, v0.7.0 expansion | `docs/rails/harness/filesystem-evidence-rail.md`, `docs/rails/harness/workflow-rail.md` |
+| Evidence Pack | Connect build/test/audit/pack/raw IDs/policy decisions/release notes | v0.5.0 seeds, v0.7.0 expansion | `docs/rails/harness/harness-rail.md`, `docs/rails/harness/workflow-rail.md` |
+| Evaluation Suite | Quality benchmarks for context, payload formats, diagrams, evidence and role packs | v0.6.1 | `docs/rails/evaluation/evaluation-suite.md`, `docs/rails/evaluation/benchmarking.md`, `ROADMAP.md` |
+| Reverse Specification Rail | Turn existing code/docs/config/logs into claims, rules, specs, gaps and validation tasks | v0.8.0 primary | `docs/rails/knowledge/reverse-specification-rail.md`, `docs/rails/knowledge/knowledge-to-rules.md` |
+| Project Brain | Verified claims, decisions, bugs, gaps, rules, stale events and agent-safe briefs | v0.8.0 | `docs/rails/knowledge/project-brain.md`, `ROADMAP.md` |
+| Host Compatibility Rail | Host-aware exports, capability matrix, host doctors and conservative prompt-only fallback for OpenCode, Antigravity, Claude, Codex, Cursor, Deep Agents-style and generic hosts | v1.1.0 implemented | `docs/rails/hosts/host-compatibility-rail.md`, `docs/rails/hosts/agents.md`, `docs/rails/hosts/mcp.md`, `docs/reference/contracts/agent-export-contract.md` |
+| Agent QA Rail | Offline datasets, golden exports, regression reports and optional LLM-as-judge policy | v1.1.1 docs, v1.3.0 planned | `docs/rails/evaluation/agent-qa-rail.md`, `docs/rails/evaluation/eval-datasets.md`, `docs/rails/evaluation/golden-agent-tests.md`, `docs/rails/evaluation/llm-as-judge-policy.md` |
+| Evidence And Provenance Rail | Provenance sidecars, verification statuses and evidence per run/report | v1.3.0 | `docs/rails/evidence/evidence-provenance-rail.md`, `docs/rails/evidence/report-rail.md`, `docs/rails/harness/filesystem-evidence-rail.md` |
+| Knowledge Rail | Compile docs/specs/notes into on-demand SKILL.md/topic/glossary/pattern packs | v1.3.0 docs, v1.4.0 skill integration | `docs/rails/knowledge/knowledge-rail.md`, `docs/rails/skills/skill-rail-2.md`, `docs/rails/knowledge/project-brain.md` |
+| Resilience Rail | Rate-limit, fallback and provider-risk documentation/reporting without proxying traffic | v1.5.0 | `docs/rails/governance/resilience-rail.md`, `docs/rails/governance/rate-limit-and-fallback-policy.md`, `docs/rails/governance/governance-cost-rail.md` |
+| Host Router Rail | Route one local context source into host-specific export formats with safe fallback | v1.5.0 | `docs/rails/hosts/host-router-rail.md`, `docs/rails/hosts/host-compatibility-rail.md`, `docs/rails/context/context-packs.md` |
+| Tasklet Rail | Small local reusable task templates for agents | v1.4.0 exploration | `docs/rails/tasklets/tasklet-rail.md`, `docs/rails/skills/skill-rail-2.md`, `docs/rails/harness/workflow-rail.md` |
+| Agent Governance/Evolution Rail | Trace, ledger, experiments, approval gates and improve/eval/apply loop | v1.6.0 | `docs/rails/governance/agent-governance-rail.md`, `docs/rails/evaluation/evaluation-suite.md`, `docs/rails/governance/policy-rail.md` |
+| Spec Rail | PRD, requirements, design, tasks and acceptance criteria as workflow inputs | v1.2.0 | `docs/rails/design/spec-driven-workflow.md`, `docs/rails/design/design-rail.md`, `docs/rails/design/diagram-rail.md` |
+| Design Rail | Local `DESIGN.md`, design token lint/diff/export and agent-readable visual guidance | v1.2.0 | `docs/rails/design/design-rail.md`, `docs/architecture/dashboard-rail.md` |
+| Knowledge Graph Rail | Local graph of files, claims, decisions, tests, workflows, diagrams and releases | v1.3.0 | `docs/rails/knowledge/knowledge-graph-rail.md`, `docs/rails/knowledge/code-graph.md`, `docs/rails/knowledge/project-brain.md` |
+| Skill Rail 2.0 | Domain skill templates, lint/eval/report and role-aware exports | v1.4.0 | `docs/rails/skills/skill-rail-2.md`, `docs/rails/skills/skill-rail.md`, `docs/rails/governance/policy-rail.md` |
+| Governance And Cost Rail | Context budget, dynamic workflow risk and MCP/skill exposure warnings | v1.5.0 | `docs/rails/governance/governance-cost-rail.md`, `docs/rails/evidence/report-rail.md`, `docs/rails/governance/policy-rail.md` |
+| Redacted Evidence Sharing | Local redacted evidence bundles for reports/logs without hosting by default | v1.4+ exploration | `docs/rails/evidence/report-redaction.md`, `docs/ecosystem/external-projects-audit.md` |
 | Local Dashboard | Local reports, trace viewer, Mermaid rendering and policy/evidence views | v0.10.0 | `ROADMAP.md` |
-| SotuRail Conductor | Optional future planner/verifier/reviewer/tasklet coordinator behind approval gates | Future proposed | `docs/conductor-mode.md`, `docs/security-boundaries.md` |
+| SotuRail Conductor | Optional future planner/verifier/reviewer/tasklet coordinator behind approval gates | Future proposed | `docs/ecosystem/conductor-mode.md`, `docs/security/security-boundaries.md` |
 
 ## Version Summary
 
@@ -254,26 +254,18 @@ Focus:
 
 ### v1.3.0
 
-Focus:
-
-- Knowledge, Evidence and Evaluation Rail;
-- graph build/explain/impact/tour;
-- Knowledge Rail ingest/estimate/compile/update/verify;
-- evidence and provenance sidecars;
-- local eval datasets, golden checks and regression reports;
-- optional LLM-as-judge outside default release gates;
-- stale-edge and missing-evidence detection.
+Absorbed into v1.4.0. Knowledge packs, evidence/provenance, deterministic evaluation, skills and tasklets share one source-backed lifecycle and are more useful when shipped together. The larger graph runtime remains planned.
 
 ### v1.4.0
 
-Focus:
+Implemented focus:
 
-- Skill Rail 2.0, Knowledge-to-Skill and Tasklet Packs;
-- domain skill templates;
-- skill lint/eval/report;
-- role-aware exports and safety gates;
-- `skill build` and `skill fold-in` planning;
-- small reusable tasklet templates.
+- Knowledge Rail estimate/compile/update/verify/list;
+- evidence collection, verification statuses and provenance reports;
+- local eval datasets, golden checks and regression reports;
+- Skill Rail 2.0 templates, build/fold-in, lint, eval and reports;
+- dry-run local tasklet templates;
+- reorganized docs and link validation.
 
 ### v1.5.0
 
@@ -319,40 +311,40 @@ SotuRail should not become:
 ## Related Docs
 
 - `ROADMAP.md`
-- `docs/roadmap-agent-runtime-addendum.md`
-- `docs/memory-rail.md`
-- `docs/harness-rail.md`
-- `docs/policy-rail.md`
-- `docs/diagram-rail.md`
-- `docs/structured-payload-rail.md`
-- `docs/agent-docs-hygiene.md`
-- `docs/filesystem-evidence-rail.md`
-- `docs/evaluation-suite.md`
-- `docs/workflow-rail.md`
-- `docs/context-packs.md`
-- `docs/spec-driven-workflow.md`
-- `docs/security-model.md`
-- `docs/rules.md`
-- `docs/benchmarking.md`
-- `docs/metrics.md`
-- `docs/mcp.md`
-- `docs/agents.md`
-- `docs/ecosystem-influences.md`
-- `docs/comparisons.md`
-- `docs/roadmap-harness-diagram-payload-addendum.md`
-- `docs/agent-harness-synthesis-2026.md`
-- `docs/agent-qa-rail.md`
-- `docs/eval-datasets.md`
-- `docs/golden-agent-tests.md`
-- `docs/llm-as-judge-policy.md`
-- `docs/evidence-provenance-rail.md`
-- `docs/agent-governance-rail.md`
-- `docs/harness-lifecycle-rail.md`
-- `docs/knowledge-rail.md`
-- `docs/resilience-rail.md`
-- `docs/rate-limit-and-fallback-policy.md`
-- `docs/multi-agent-workflow-templates.md`
-- `docs/host-router-rail.md`
-- `docs/tasklet-rail.md`
-- `docs/security-boundaries.md`
-- `docs/conductor-mode.md`
+- `docs/roadmap/roadmap-agent-runtime-addendum.md`
+- `docs/rails/context/memory-rail.md`
+- `docs/rails/harness/harness-rail.md`
+- `docs/rails/governance/policy-rail.md`
+- `docs/rails/design/diagram-rail.md`
+- `docs/rails/context/structured-payload-rail.md`
+- `docs/rails/hosts/agent-docs-hygiene.md`
+- `docs/rails/harness/filesystem-evidence-rail.md`
+- `docs/rails/evaluation/evaluation-suite.md`
+- `docs/rails/harness/workflow-rail.md`
+- `docs/rails/context/context-packs.md`
+- `docs/rails/design/spec-driven-workflow.md`
+- `docs/security/security-model.md`
+- `docs/rails/governance/rules.md`
+- `docs/rails/evaluation/benchmarking.md`
+- `docs/rails/evaluation/metrics.md`
+- `docs/rails/hosts/mcp.md`
+- `docs/rails/hosts/agents.md`
+- `docs/ecosystem/ecosystem-influences.md`
+- `docs/ecosystem/comparisons.md`
+- `docs/roadmap/roadmap-harness-diagram-payload-addendum.md`
+- `docs/ecosystem/agent-harness-synthesis-2026.md`
+- `docs/rails/evaluation/agent-qa-rail.md`
+- `docs/rails/evaluation/eval-datasets.md`
+- `docs/rails/evaluation/golden-agent-tests.md`
+- `docs/rails/evaluation/llm-as-judge-policy.md`
+- `docs/rails/evidence/evidence-provenance-rail.md`
+- `docs/rails/governance/agent-governance-rail.md`
+- `docs/rails/harness/harness-lifecycle-rail.md`
+- `docs/rails/knowledge/knowledge-rail.md`
+- `docs/rails/governance/resilience-rail.md`
+- `docs/rails/governance/rate-limit-and-fallback-policy.md`
+- `docs/rails/tasklets/multi-agent-workflow-templates.md`
+- `docs/rails/hosts/host-router-rail.md`
+- `docs/rails/tasklets/tasklet-rail.md`
+- `docs/security/security-boundaries.md`
+- `docs/ecosystem/conductor-mode.md`

package/docs/{repo-docs-audit-2026-06-05.md → roadmap/repo-docs-audit-2026-06-05.md}

@@ -7,7 +7,7 @@ This audit records the documentation consistency check after the 2026 agent-harn
 Checked areas:
 
 - main docs entry points: `README.md`, `ROADMAP.md`, `CHANGELOG.md`;
-- planning parents: `docs/future-rails-index.md`, `docs/ecosystem-influences.md`, `docs/external-projects-audit.md`;
+- planning parents: `docs/roadmap/future-rails-index.md`, `docs/ecosystem/ecosystem-influences.md`, `docs/ecosystem/external-projects-audit.md`;
 - old related docs: workflow, context packs, report, observability, host docs, evaluation, harness, skills and governance;
 - new future rails: Agent QA, Evidence/Provenance, Agent Governance, Harness Lifecycle, Knowledge, Resilience, Host Router and Tasklet;
 - examples under `examples/workflows/`.
@@ -18,15 +18,15 @@ The 2026 ideas are now represented as documentation and roadmap items, not as im
 
 Confirmed coverage:
 
-- QA/eval/golden checks are covered by `docs/agent-qa-rail.md`, `docs/eval-datasets.md`, `docs/golden-agent-tests.md` and `docs/llm-as-judge-policy.md`.
-- Validate/fix/verify/report pipeline ideas are covered by `docs/observability-rail.md`, `docs/workflow-rail.md` and `examples/workflows/agent-pipeline-workflow.md`.
-- OpenTracy-style trace, ledger, boundary, experiments and approval ideas are covered by `docs/agent-governance-rail.md`.
-- Harness engineering lifecycle is covered by `docs/harness-lifecycle-rail.md` and linked from `docs/harness-rail.md` and `docs/workflow-rail.md`.
-- ECC-style host/profile/doctor/audit/skills ideas are covered across `docs/future-rails-index.md`, `docs/host-compatibility-rail.md`, `docs/agent-hosts.md` and `docs/skill-rail-2.md`.
-- Feynman-style provenance is covered by `docs/evidence-provenance-rail.md` and linked from Report and Workflow docs.
-- book-to-skill-style document-to-skill packs are covered by `docs/knowledge-rail.md`, `docs/context-packs.md` and `docs/skill-rail-2.md`.
-- Tasklet ideas are covered by `docs/tasklet-rail.md` and linked from Context/Workflow/Skill planning docs.
-- 9Router ideas are safely mapped to `docs/host-router-rail.md`, without proxy, MITM, quota bypass or credential-routing scope.
+- QA/eval/golden checks are covered by `docs/rails/evaluation/agent-qa-rail.md`, `docs/rails/evaluation/eval-datasets.md`, `docs/rails/evaluation/golden-agent-tests.md` and `docs/rails/evaluation/llm-as-judge-policy.md`.
+- Validate/fix/verify/report pipeline ideas are covered by `docs/architecture/observability-rail.md`, `docs/rails/harness/workflow-rail.md` and `examples/workflows/agent-pipeline-workflow.md`.
+- OpenTracy-style trace, ledger, boundary, experiments and approval ideas are covered by `docs/rails/governance/agent-governance-rail.md`.
+- Harness engineering lifecycle is covered by `docs/rails/harness/harness-lifecycle-rail.md` and linked from `docs/rails/harness/harness-rail.md` and `docs/rails/harness/workflow-rail.md`.
+- ECC-style host/profile/doctor/audit/skills ideas are covered across `docs/roadmap/future-rails-index.md`, `docs/rails/hosts/host-compatibility-rail.md`, `docs/rails/hosts/agent-hosts.md` and `docs/rails/skills/skill-rail-2.md`.
+- Feynman-style provenance is covered by `docs/rails/evidence/evidence-provenance-rail.md` and linked from Report and Workflow docs.
+- book-to-skill-style document-to-skill packs are covered by `docs/rails/knowledge/knowledge-rail.md`, `docs/rails/context/context-packs.md` and `docs/rails/skills/skill-rail-2.md`.
+- Tasklet ideas are covered by `docs/rails/tasklets/tasklet-rail.md` and linked from Context/Workflow/Skill planning docs.
+- 9Router ideas are safely mapped to `docs/rails/hosts/host-router-rail.md`, without proxy, MITM, quota bypass or credential-routing scope.
 
 ## Link Check