mdkg 0.3.6 → 0.3.8

package/dist/init/AGENT_START.md

@@ -29,8 +29,8 @@ Agent operating prompt:
 - Treat goal `required_checks` as report-only guidance from mdkg. Run commands yourself, then record evidence in the goal or active work item.
 - Record skill improvement candidates during normal goal execution; edit `SKILL.md` only when the active node is explicit skill-maintenance work.
 - Use `mdkg skill list`, `mdkg skill search`, and `mdkg skill show <slug>` for skill discovery.
-- Use `mdkg capability list/search/show` for deterministic skills, `SPEC.md`, `WORK.md`, core-doc, and design-doc capability discovery.
-- Use `mdkg spec list/show/validate` for focused optional `SPEC.md` capability records.
+- Use `mdkg capability list/search/show` for deterministic skills, `MANIFEST.md` / legacy `SPEC.md`, `WORK.md`, core-doc, and design-doc capability discovery.
+- Use `mdkg manifest list/show/validate` for focused optional `MANIFEST.md` capability records; `mdkg spec ...` remains a legacy alias for one compatibility release.
 - Use `mdkg index` to refresh JSON compatibility caches and `.mdkg/index/mdkg.sqlite` when SQLite mode is enabled.
 - Treat `.mdkg/db` as project application state; use `mdkg db init` to create
   the generic scaffold and enable `db.enabled` without creating an active
@@ -39,7 +39,10 @@ Agent operating prompt:
   node:sqlite queue delivery, internal event/receipt/reducer, writer lease/CAS,
   and queue control migrations. Queue state is delivery infrastructure, not
   canonical event history; use `mdkg db queue ...` to create, pause, enqueue,
-  claim, settle, inspect, and drain local queues. Event rows are durable local
+  claim, settle, inspect, and drain local queues. Use
+  `mdkg db queue contract --json` for the public adapter contract covering
+  payload hashing, dedupe, claim order, lease-owner settlement, retry,
+  dead-letter, release-expired, pause/resume, snapshot policy, and stats. Event rows are durable local
   project DB history; receipts, reducers, writer leases, and materializers are
   internal local helper surfaces, with no public `mdkg db event`,
   `mdkg db reducer`, `mdkg db lease`, or `mdkg db materializer` CLI yet. Use `mdkg db verify` and `mdkg db stats` for
@@ -108,9 +111,9 @@ Capability discovery:
 - `mdkg capability list --kind skill --json`
 - `mdkg capability search "<query>" --kind spec --json`
 - `mdkg capability search "<query>" --kind work --json`
-- `mdkg spec list --json`
-- `mdkg spec show <id-or-qid-or-alias> --json`
-- `mdkg spec validate <id-or-qid-or-alias> --json`
+- `mdkg manifest list --json`
+- `mdkg manifest show <id-or-qid-or-alias> --json`
+- `mdkg manifest validate <id-or-qid-or-alias> --json`
 
 Conventions:
 - `AGENTS.md` is the Codex/OpenAI-oriented wrapper doc.

package/dist/init/CLI_COMMAND_MATRIX.md

@@ -6,8 +6,9 @@ Verify live help with:
 - `mdkg --help`
 - `mdkg help <command>`
 
-Optional reusable SPEC capability records are accessed through `mdkg spec ...`.
-Repos without SPEC files remain valid.
+Optional reusable manifest capability records are accessed through `mdkg manifest ...`.
+Repos without MANIFEST/SPEC files remain valid. `mdkg spec ...` remains a
+legacy alias for one compatibility release.
 
 Primary commands:
 - `mdkg init`
@@ -17,8 +18,10 @@ Primary commands:
 - `mdkg list`
 - `mdkg search`
 - `mdkg pack`
+- `mdkg handoff create <id-or-qid> [--ws <alias>] [--depth <n>] [--out <path>] [--json]`
 - `mdkg skill`
 - `mdkg capability`
+- `mdkg manifest`
 - `mdkg spec`
 - `mdkg archive`
 - `mdkg bundle`
@@ -34,6 +37,14 @@ Primary commands:
 - `mdkg fix apply [--family ids] [--target <id-or-qid>] [--base-ref <ref>] [--json]`
 - `mdkg fix ids [--target <id-or-qid>] [--base-ref <ref>] [--apply] [--json]`
 
+Semantic refs:
+- work nodes may use `context_refs` for non-executable background and `evidence_refs` for proof/audit refs
+- semantic refs may point at local ids, subgraph qids, or URI refs
+- use `mdkg pack <id> --edges context_refs,evidence_refs` when a handoff should traverse semantic refs
+- use `mdkg handoff create <id-or-qid> --json` for sanitized copy-ready agent handoff prompts
+- handoffs summarize graph state, latest checkpoint, boundaries, required checks, next actions, and raw-marker warnings without copying raw node bodies
+- executable goal queues still belong in goal `scope_refs`
+
 Operator health:
 - `mdkg status [--json]` is a read-only summary for scripts and agents
 - reports mdkg version/config, git state, graph/index freshness, selected-goal state, project DB verification summary, and generated cache status
@@ -71,6 +82,7 @@ Project database commands:
 - `mdkg db verify [--json]`
 - `mdkg db stats [--json]`
 - `mdkg db queue create|pause|resume|enqueue|claim|ack|fail|dead-letter|release-expired|stats|list|show ... [--json]`
+- `mdkg db queue contract [--json]`
 - `mdkg db snapshot seal [--queue-policy drain|paused] [--json]`
 - `mdkg db snapshot verify [--json]`
 - `mdkg db snapshot status [--json]`
@@ -87,6 +99,9 @@ Project database commands:
   in the configured migration table
 - `mdkg db queue ...` exposes durable local delivery operations backed by
   node:sqlite; queue rows are delivery state, not canonical event history
+- `mdkg db queue contract --json` returns the public adapter contract for
+  payload hashes, dedupe, claim order, lease-owner settlement, retries,
+  dead-letter, release-expired, pause/resume, snapshot policy, and stats
 - paused queues reject enqueue and claim, but ack/fail/dead-letter and
   release-expired remain available so leased work can settle
 - event tables are durable local history for project DB state transitions;
@@ -105,7 +120,11 @@ Project database commands:
 - active `.mdkg/db/runtime/` files and `.mdkg/db` WAL/SHM/journal/lock/temp files are ignored by default
 
 Validation commands:
-- `mdkg validate [--out <path>] [--quiet] [--json]`
+- `mdkg validate [--out <path>] [--json-out <path>] [--quiet] [--changed-only] [--summary] [--limit <n>] [--json]`
+- `--changed-only` filters warning presentation to changed `.mdkg` files while full graph errors still run
+- `--summary` emits bounded warning samples for agent/CI logs; `--limit <n>` controls the sample size
+- `--out <path>` writes the compatibility text report; `--json-out <path>` writes a clean full JSON receipt
+- JSON receipts include `warning_summary` and `warning_diagnostics` with warning ids, categories, severity, paths, refs, and remediation text
 
 Node creation commands:
 - `mdkg new <type> "<title>" [options] [--json]`
@@ -113,19 +132,21 @@ Node creation commands:
 - `mdkg new spike "<research question>" [options] [--json]`
 
 Agent workflow file type creation:
-- `mdkg new spec "<title>" [options] [--json]`
+- `mdkg new manifest "<title>" [options] [--json]`
 - `mdkg new work "<title>" [options] [--json]`
 - `mdkg new work_order "<title>" [options] [--json]`
 - `mdkg new receipt "<title>" [options] [--json]`
 - `mdkg new feedback "<title>" [options] [--json]`
 - `mdkg new dispute "<title>" [options] [--json]`
 - `mdkg new proposal "<title>" [options] [--json]`
-- `mdkg new spec "image worker" --id agent.image-worker`
+- `mdkg new spec "<title>" [options] [--json]` is a deprecated alias for
+  `mdkg new manifest` and creates `MANIFEST.md` with `type: manifest`
+- `mdkg new manifest "image worker" --id agent.image-worker`
 - `mdkg new work "generate image" --id work.generate-image`
 
 Agent workflow notes:
 - `--id <portable-id>` is only for agent workflow file types.
-- `spec` and `work` scaffold as validation-clean standalone docs.
+- `manifest` and `work` scaffold as validation-clean standalone docs.
 - `work_order`, `receipt`, `feedback`, `dispute`, and `proposal` need real refs before strict `mdkg validate` passes.
 - `goal` nodes capture recursive objective state and required checks, but normal `mdkg next` does not select them.
 - `spike` nodes are actionable research/planning work under `.mdkg/work/`; use `mdkg task start|update|done` for lifecycle state.
@@ -146,11 +167,12 @@ Event log commands:
 Task mutation commands:
 - `mdkg task start <id-or-qid> [--ws <alias>] [--run-id <id>] [--note "<text>"] [--json]`
 - `mdkg task update <id-or-qid> [options] [--json]`
-- `mdkg task done <id-or-qid> [--checkpoint "<title>"] [options] [--json]`
+- `mdkg task done <id-or-qid> [--checkpoint "<title>"] [--checkpoint-kind implementation|test-proof|goal-closeout|audit|handoff] [options] [--json]`
 - task commands support task-like `feat`, `task`, `bug`, `test`, and `spike` nodes
 
 Checkpoint commands:
-- `mdkg checkpoint new <title> [--ws <alias>] [--json]`
+- `mdkg checkpoint new <title> [--kind implementation|test-proof|goal-closeout|audit|handoff] [--ws <alias>] [--json]`
+- checkpoint kinds render bodies with command evidence, pass/fail status, known warnings, changed surfaces, boundaries, and follow-up refs
 
 Agent bootstrap:
 - `mdkg init --agent`
@@ -179,18 +201,25 @@ Capability discovery:
 - `mdkg capability resolve [query] [--requires <capability>] [--fresh-only] [--json]`
 - capability records are deterministic cache projections from Markdown
 - records include source hash, headings, refs, and `indexed_at`
-- SPEC and WORK capability records include read-only `linkage` arrays for related SPECs, work contracts, work orders, and receipts when those graph mirrors exist
+- MANIFEST/SPEC and WORK capability records include read-only `linkage` arrays for related manifests, work contracts, work orders, and receipts when those graph mirrors exist
 - normal task, epic, feat, bug, test, spike, and checkpoint nodes are intentionally excluded
 
-Spec capability records:
+Manifest capability records:
+- `mdkg manifest list [--json]`
+- `mdkg manifest show <id-or-qid-or-alias> [--json]`
+- `mdkg manifest validate [<id-or-qid-or-alias>] [--json]`
+- `MANIFEST.md` is optional; repos with no manifest files still validate
+- manifest records describe reusable capability surfaces, not general planning notes
+- `mdkg manifest validate` with no ref validates the graph and all optional manifest records
+- `mdkg manifest validate <ref>` also checks that the target manifest reference exists
+- `mdkg manifest ...` is the focused manifest command family; `mdkg capability ...` remains broader skill/manifest/work/core/design discovery
+
+Legacy spec capability records:
 - `mdkg spec list [--json]`
 - `mdkg spec show <id-or-qid-or-alias> [--json]`
 - `mdkg spec validate [<id-or-qid-or-alias>] [--json]`
-- `SPEC.md` is optional; repos with no SPEC files still validate
-- SPEC records describe reusable capability surfaces, not general planning notes
-- `mdkg spec validate` with no ref validates the graph and all optional SPEC records
-- `mdkg spec validate <ref>` also checks that the target SPEC reference exists
-- `mdkg spec ...` is the focused SPEC command family; `mdkg capability ...` remains broader skill/spec/work/core/design discovery
+- `mdkg spec ...` is a deprecated alias for `mdkg manifest ...` during the compatibility bridge
+- legacy `SPEC.md` files remain valid for one compatibility release
 
 Archive sidecars:
 - `mdkg archive add <file> [--id <archive.id>] [--kind source|artifact] [--visibility private|internal|public] [--title <title>] [--refs <...>] [--relates <...>] [--json]`
@@ -220,6 +249,7 @@ Graph clone, fork, and template import:
 - `mdkg graph clone <source-bundle-or-mdkg-dir> --target <path> [--json]`
 - `mdkg graph fork <source-bundle-or-mdkg-dir> --target <path> [--start-goal <goal-id>] [--json]`
 - `mdkg graph import-template <source-bundle-or-mdkg-dir> [--start-goal <goal-id>] [--select-goal] [--id-prefix <prefix>] [--dry-run] [--apply] [--json]`
+- `mdkg graph refs <id-or-qid> [--ws <alias>] [--json]`
 - `graph clone` and `graph fork` preserve IDs because the target is a separate graph namespace
 - clone/fork targets must be empty or absent and stay under the current mdkg root
 - live directory sources are never mutated; clone/fork refuses targets nested inside a live source directory
@@ -230,6 +260,7 @@ Graph clone, fork, and template import:
 - colliding semantic template IDs require `--id-prefix`
 - `--select-goal` requires `--start-goal`; on apply it activates the imported start goal, pauses competing active root goals, validates, then writes selected-goal state
 - importing active template goals without `--select-goal` fails before writing when it would create multiple active root goals
+- `graph refs` is read-only and summarizes inbound/outbound scope, context, evidence, blocker, related, and structural refs across local and subgraph qids
 - subgraphs remain read-only bundle projections for orchestration context; use `graph clone|fork|import-template` when authored graph state should be created
 
 Subgraph orchestration:
@@ -263,11 +294,13 @@ Work semantic mirrors:
 - `mdkg work receipt verify <id-or-qid> [--json]`
 - `mdkg work receipt update <id-or-qid> [--receipt-status <status>] [--add-artifacts <...>] [--add-proof-refs <...>] [--add-attestation-refs <...>] [--json]`
 - `mdkg work artifact add <order-or-receipt-id-or-qid> <file> [--id <archive.id>] [--kind source|artifact] [--json]`
-- `work trigger` accepts a `WORK.md` ref directly or a `SPEC.md` capability ref with exactly one resolvable work contract; it creates a submitted order mirror and never executes work
+- `mdkg work validate [<id-or-qid>] [--type manifest|spec|work|work_order|receipt|feedback|dispute|proposal] [--json]`
+- `work trigger` accepts a `WORK.md` ref directly or a `MANIFEST.md` / legacy `SPEC.md` capability ref with exactly one resolvable work contract; it creates a submitted order mirror and never executes work
 - example: `mdkg work trigger work.example --id order.example-1 --requester user://example --json`
 - `work trigger --enqueue <queue>` requires a valid project DB plus an explicitly created active queue, creates a submitted order mirror, and enqueues a local delivery message without executing work
 - `work order status` is read-only and reports deterministic order state plus linked receipts
 - `work receipt verify` is read-only and reports linkage, evidence, archive ref, hash, outcome, and redaction-policy checks
+- `work validate` is read-only and reports typed diagnostics for agent workflow mirrors; obvious raw secret, prompt, token, or payload markers are warnings, not hard failures
 - work commands mutate mdkg semantic mirror files only; production order, receipt, feedback, dispute, payment, ledger, marketplace inventory, fulfillment, and execution state remains canonical outside mdkg
 - do not store raw secrets, credentials, live payment state, ledger mutations, or canonical marketplace state in work mirrors
 - `artifact://...` refs identify external/runtime-managed artifacts; `archive://...` refs identify committed mdkg archive sidecars
@@ -299,6 +332,7 @@ Goal nodes:
 - `goal activate` makes one local root goal active, pauses competing local active goals in the same workspace, and writes selected-goal state
 - `goal archive` marks a superseded historical goal archived so it remains readable but not actionable
 - `goal next` is read-only; use `goal claim` to set `active_node`
+- `goal done` preserves the final actionable item as `last_active_node` and removes actionable `active_node`
 - `mdkg goal evaluate` is report-only and never runs commands from `required_checks`
 - skill improvements discovered during normal goal execution should be recorded as candidates or proposals unless the active node is skill-maintenance
 

package/dist/init/README.md

@@ -25,8 +25,9 @@ mdkg new task "..." --status todo --priority 1
 mdkg search "..."
 mdkg show <id>
 mdkg pack <id>
+mdkg handoff create <id-or-qid> --json
 mdkg capability search "..."
-mdkg spec list --json
+mdkg manifest list --json
 mdkg archive list
 mdkg bundle create --profile private
 mdkg graph clone .mdkg/bundles/private/all.mdkg.zip --target demos/demo-1 --json
@@ -36,7 +37,14 @@ mdkg fix plan --json
 mdkg validate
 ```
 
-This repo is already initialized. Use `mdkg upgrade` to preview safe scaffold updates, `mdkg index` to create or refresh generated graph/skill/capability/subgraph and SQLite caches after init, `mdkg new` to create work, `mdkg new goal "..."` plus `mdkg goal activate/current/next/claim/evaluate` for recursive long-running objectives, `mdkg search`/`mdkg show` to inspect graph state, `mdkg capability ...` to inspect cached skill/spec/work/core/design capabilities, `mdkg spec ...` for focused optional SPEC records, `mdkg capability resolve ...` to rank local and subgraph capabilities, `mdkg archive ...` to register source/artifact sidecars, `mdkg work ...` to create work contract/order/receipt semantic mirrors and deterministic trigger/verification records, `mdkg bundle ...` to create full graph snapshot bundles, `mdkg graph ...` to clone/fork/import authored graph templates, `mdkg subgraph ...` to register read-only child graph planning views, `mdkg pack <id>` to build deterministic context, and `mdkg validate` before closeout.
+This repo is already initialized. Use `mdkg upgrade` to preview safe scaffold updates, `mdkg index` to create or refresh generated graph/skill/capability/subgraph and SQLite caches after init, `mdkg new` to create work, `mdkg new goal "..."` plus `mdkg goal activate/current/next/claim/evaluate` for recursive long-running objectives, `mdkg search`/`mdkg show` to inspect graph state, `mdkg capability ...` to inspect cached skill/manifest/spec/work/core/design capabilities, `mdkg manifest ...` for focused optional manifest records, `mdkg capability resolve ...` to rank local and subgraph capabilities, `mdkg archive ...` to register source/artifact sidecars, `mdkg work ...` to create work contract/order/receipt semantic mirrors and deterministic trigger/verification records, `mdkg bundle ...` to create full graph snapshot bundles, `mdkg graph ...` to clone/fork/import authored graph templates, `mdkg subgraph ...` to register read-only child graph planning views, `mdkg pack <id>` to build deterministic context, `mdkg handoff create <id-or-qid> --json` to create a sanitized copy-ready agent handoff prompt, and `mdkg validate` before closeout.
+
+`mdkg handoff create` summarizes goal/work state, included pack nodes, latest
+checkpoint, boundaries, required checks, next actions, and raw-content marker
+warnings without copying raw node bodies into the prompt. Use `--out` to write
+the handoff artifact inside the repo root.
+
+For large historical graphs, `mdkg validate --changed-only --json` keeps warning review focused on changed `.mdkg` files while full graph errors still run. `mdkg validate --summary --json --limit 20` keeps agent and CI logs bounded, and `--json-out <path>` writes a clean full JSON receipt artifact. `mdkg format --headings --dry-run --summary --json --limit 20` previews missing recommended heading additions with bounded output before `--apply`.
 
 Use `mdkg status --json` for a read-only operator summary of Git, graph,
 selected-goal, project DB, and generated-cache health before mutating work. Use
@@ -72,15 +80,17 @@ intentionally with `mdkg new task ...` or `mdkg new test ...`.
 Agent workflow docs can use semantic ids:
 
 ```bash
-mdkg new spec "image worker" --id agent.image-worker
+mdkg new manifest "image worker" --id agent.image-worker
 mdkg new work "generate image" --id work.generate-image
 ```
 
-`SPEC.md` is optional. Repos without SPEC files still validate. When present,
-SPEC records describe reusable capability surfaces rather than general planning
-notes. `mdkg spec list/show/validate` is the focused SPEC command family, while
-`mdkg capability ...` remains the broader read-only discovery surface for
-skills, SPECs, WORK contracts, core docs, and design docs.
+`MANIFEST.md` is optional. Repos without manifest files still validate. When
+present, manifest records describe reusable capability surfaces rather than
+general planning notes. `SPEC.md` remains a legacy alias for one compatibility
+release, and `mdkg spec list/show/validate` remains a deprecated alias for
+`mdkg manifest list/show/validate`. `mdkg capability ...` remains the broader
+read-only discovery surface for skills, manifests, WORK contracts, core docs,
+and design docs.
 
 Read `AGENT_START.md` first when this repo includes it.
 
@@ -124,7 +134,11 @@ infrastructure, not canonical event history; use `mdkg db queue ...` to create,
 pause, enqueue, claim, settle, inspect, and drain local queues. Event rows are
 durable local project DB history; receipts, reducers, writer leases, and
 materializers are internal local helper surfaces, with no public `mdkg db event`,
-`mdkg db reducer`, `mdkg db lease`, or `mdkg db materializer` CLI yet. Use `mdkg db verify` for non-mutating health checks and
+`mdkg db reducer`, `mdkg db lease`, or `mdkg db materializer` CLI yet.
+`mdkg db queue contract --json` returns the read-only public adapter contract
+for canonical payload hashing, dedupe keys, oldest-ready claim order,
+lease-owner checked settlement, retry/dead-letter behavior, release-expired,
+pause/resume, snapshot queue policy, and stats. Use `mdkg db verify` for non-mutating health checks and
 `mdkg db stats` for table counts, DB size, migration state, and receipt-file
 counts. Use `mdkg db snapshot seal` to create an opt-in sealed checkpoint under
 `.mdkg/db/state`; the default queue policy is drain, and
@@ -235,6 +249,9 @@ Update and artifact commands accept local ids or local qids; subgraph qids are r
 `mdkg work trigger` creates a deterministic submitted `WORK_ORDER.md` from a
 WORK contract or a SPEC with exactly one resolvable work contract. `mdkg work
 order status` and `mdkg work receipt verify` are read-only review helpers.
+`mdkg work validate [<id-or-qid>] [--type spec|work|work_order|receipt|feedback|dispute|proposal] --json`
+is a read-only focused validator for agent workflow mirrors with typed diagnostics
+and raw secret, prompt, token, or payload marker warnings.
 `mdkg work trigger --enqueue <queue>` optionally writes a local project DB queue
 delivery message after the queue has been explicitly created and is active; it
 still does not execute work.

package/dist/init/init-manifest.json

@@ -1,7 +1,7 @@
 {
   "schema_version": 1,
   "tool": "mdkg",
-  "mdkg_version": "0.3.6",
+  "mdkg_version": "0.3.8",
   "files": [
     {
       "path": ".mdkg/config.json",
@@ -61,7 +61,7 @@
     {
       "path": ".mdkg/README.md",
       "category": "mdkg_doc",
-      "sha256": "cecb50528b70c0c4e32f8a052a668b5d87d6ac542818229efa848a6b1ac42d62"
+      "sha256": "e876acfbf3e9de411cd45eb86e7e9bff821f8a5dd6dc933bd6da4ac88a9686e6"
     },
     {
       "path": ".mdkg/skills/build-pack-and-execute-task/SKILL.md",
@@ -91,12 +91,12 @@
     {
       "path": ".mdkg/templates/default/bug.md",
       "category": "template",
-      "sha256": "88e3f8d67aa3b2156be2006bd1c8139b7b16b46f2c05a4f4d840b3746a40f9c5"
+      "sha256": "27ed977c67030befccfc469d9db257fadc9ad5331963e259c489ba03754194cb"
     },
     {
       "path": ".mdkg/templates/default/chk.md",
       "category": "template",
-      "sha256": "15f793e422009c76f8bda72c3609157acd7f03762d9c9bdb096c4aa2e2c3fb89"
+      "sha256": "9e32f46665fc7cc582c4695772f5d99799f574229e740fa8892c7b9e75324de9"
     },
     {
       "path": ".mdkg/templates/default/dec.md",
@@ -116,12 +116,12 @@
     {
       "path": ".mdkg/templates/default/epic.md",
       "category": "template",
-      "sha256": "56697b69af027fe08c00866c0c0670f7d7e2a0857803bd4aabb99f4761311879"
+      "sha256": "0f515c046064771e3f9abaa3689634bfae3ffcf2ffe9825e8e3b7dc15c24fc50"
     },
     {
       "path": ".mdkg/templates/default/feat.md",
       "category": "template",
-      "sha256": "99cfdd97b40b119cf899fd766d816c07b7daaafd8ec8e70e6c6de5c0d37f581c"
+      "sha256": "25c04ce41d8cf519d5c3009219eeed48f76c23600c6b7999264831aa92cefc4b"
     },
     {
       "path": ".mdkg/templates/default/feedback.md",
@@ -131,7 +131,12 @@
     {
       "path": ".mdkg/templates/default/goal.md",
       "category": "template",
-      "sha256": "710252d8a2dc35be71661d6988007af127052fa8ad24b1fb00374c975ae117a2"
+      "sha256": "8f984580aefd02b34639fa7f5d2a834662656bbf2e12f14a285f5ae31aff74ce"
+    },
+    {
+      "path": ".mdkg/templates/default/manifest.md",
+      "category": "template",
+      "sha256": "2ff2af58cae6a0381a66949d7d9e997fac6486da85171b09096c43663889da95"
     },
     {
       "path": ".mdkg/templates/default/prd.md",
@@ -166,17 +171,17 @@
     {
       "path": ".mdkg/templates/default/spike.md",
       "category": "template",
-      "sha256": "ac805dca7e6edcdad35e24b615dc7399cbae3bed676b9da57e24a393eb425245"
+      "sha256": "56e881fad20bd4fda8ae53cc78ce6f30f4dacd083f4eda40f9f2ba3723d52e0e"
     },
     {
       "path": ".mdkg/templates/default/task.md",
       "category": "template",
-      "sha256": "f6385dd4f56f1c8df164ac3e79d82c5a3fde3f952dc9cdc78c0b8077caa62277"
+      "sha256": "2db6e56298b22e9da9121e2ac31f2fd68e370211218c5bcaae1014a9e63b7cde"
     },
     {
       "path": ".mdkg/templates/default/test.md",
       "category": "template",
-      "sha256": "bdae53e5cb4bd77712da155bbcfc24409ced8162c977e6bc478a9d1e695ddef7"
+      "sha256": "59484d4cf397707d738ccbc2f788c08eab58d3d73f37591f7bde8f1533b828ce"
     },
     {
       "path": ".mdkg/templates/default/work_order.md",
@@ -193,51 +198,101 @@
       "category": "template",
       "sha256": "08a1bd65297173a1dc9df95776775d406337a419d4bc51863593b6f28777ebdb"
     },
+    {
+      "path": ".mdkg/templates/specs/agent.MANIFEST.md",
+      "category": "template",
+      "sha256": "79fb755c95747d3f673e1333ec3f039d08f76488ba268b7fa32f457d8638ee64"
+    },
     {
       "path": ".mdkg/templates/specs/agent.SPEC.md",
       "category": "template",
       "sha256": "dab10c0ed12aa10a752ee3bd61f263065644826eb950c71a9e3458673edb0ca5"
     },
+    {
+      "path": ".mdkg/templates/specs/api.MANIFEST.md",
+      "category": "template",
+      "sha256": "dd342ab2bd331b716e4b7de9f8c7cfa503703422a4cc88c21fe90ff49b11aa83"
+    },
     {
       "path": ".mdkg/templates/specs/api.SPEC.md",
       "category": "template",
       "sha256": "aee86cadcca31a5a015d7e15ad7503c4aa30f2af0079ec03f857b82b3ecbae59"
     },
+    {
+      "path": ".mdkg/templates/specs/base.MANIFEST.md",
+      "category": "template",
+      "sha256": "056c4ccf8285db72c8befdc6acd165d625dbe892d70e89ce586f081db8e250fc"
+    },
     {
       "path": ".mdkg/templates/specs/base.SPEC.md",
       "category": "template",
       "sha256": "6d4171fac00c2f3d8f2a2ac746b8a47c59aaecebe224c3a0046dd6e6974a1d08"
     },
+    {
+      "path": ".mdkg/templates/specs/capability.MANIFEST.md",
+      "category": "template",
+      "sha256": "0ab2c2f90c6bdcddaafd77ac6cc36422d6ee10cc935942d8f80728f5b184d5cc"
+    },
     {
       "path": ".mdkg/templates/specs/capability.SPEC.md",
       "category": "template",
       "sha256": "68a91e8bbd80d1ff1972e4c31e29f26451d5a1be1d25d414170fdd670010066f"
     },
+    {
+      "path": ".mdkg/templates/specs/integration.MANIFEST.md",
+      "category": "template",
+      "sha256": "7af890bdac5515600661c241dd488c2bc270248ae6fba9afb0307d7a643022a4"
+    },
     {
       "path": ".mdkg/templates/specs/integration.SPEC.md",
       "category": "template",
       "sha256": "e907ce6ebc1fa5a455e31e39036e3f8699dccb3d9e45288c8ea025eaec4ca4a2"
     },
+    {
+      "path": ".mdkg/templates/specs/model.MANIFEST.md",
+      "category": "template",
+      "sha256": "3fe36a424b6eba8542bd08d3e2728ffd7f35d2c72037f2e565757cea15da2e1f"
+    },
     {
       "path": ".mdkg/templates/specs/model.SPEC.md",
       "category": "template",
       "sha256": "56061a241819dfda4d3022c075f744cf6650f5f52c58cd15b0af9d1f613af4f2"
     },
+    {
+      "path": ".mdkg/templates/specs/project.MANIFEST.md",
+      "category": "template",
+      "sha256": "e1f68306a9cff3d73e65a2fea38d52b49dd76921b5d7c7ea9710eb2c23cb6f89"
+    },
     {
       "path": ".mdkg/templates/specs/project.SPEC.md",
       "category": "template",
       "sha256": "386c41852cbb46e7a6ba583a7b0c4126262a56618d8e214aaa601b68d55818b9"
     },
+    {
+      "path": ".mdkg/templates/specs/runtime-agent.MANIFEST.md",
+      "category": "template",
+      "sha256": "4ba2013362597f101fb225bc6a3369a47a0c1cac3de78220efea4e53d42f0393"
+    },
     {
       "path": ".mdkg/templates/specs/runtime-agent.SPEC.md",
       "category": "template",
       "sha256": "53af7c3e172f5ed1297f340aca0be5e53302613d2e6bb9145915067d7b0004c8"
     },
+    {
+      "path": ".mdkg/templates/specs/runtime-image.MANIFEST.md",
+      "category": "template",
+      "sha256": "098debed22f274c9f408defcf330ef27c760e1a3166876f2a6e16cb28cfb4757"
+    },
     {
       "path": ".mdkg/templates/specs/runtime-image.SPEC.md",
       "category": "template",
       "sha256": "37416b045cd7733d1f5e1cc629ac9b6616024d5fa52f2bdcd90110267151e593"
     },
+    {
+      "path": ".mdkg/templates/specs/tool.MANIFEST.md",
+      "category": "template",
+      "sha256": "79e45600a3b83d70e2791cbdff0138b57018a30b7e277d9017b7ba142deccb91"
+    },
     {
       "path": ".mdkg/templates/specs/tool.SPEC.md",
       "category": "template",
@@ -246,7 +301,7 @@
     {
       "path": "AGENT_START.md",
       "category": "startup_doc",
-      "sha256": "46e734b6eae92eb957c3e29fb4d206e4be3fd6d32dd00e888e01a0dc75979428"
+      "sha256": "0d3bb2f9ea156f53c1bf68595e591578caadc18d296fa919e71ea1a3d9a968e8"
     },
     {
       "path": "AGENTS.md",
@@ -261,7 +316,7 @@
     {
       "path": "CLI_COMMAND_MATRIX.md",
       "category": "startup_doc",
-      "sha256": "5321d82b594c1c0aa5574dcffbf21e5a1d3df89566b0f9acfc854fcaac3e13a7"
+      "sha256": "8f426fd8da402b4b550b71e7ec3882c614ceb2d79d23c47b9c27bca7e39fadb2"
     },
     {
       "path": "llms.txt",

package/dist/init/templates/default/bug.md

@@ -16,6 +16,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 created: {{created}}

package/dist/init/templates/default/chk.md

@@ -2,6 +2,7 @@
 id: {{id}}
 type: checkpoint
 title: {{title}}
+checkpoint_kind: {{checkpoint_kind}}
 status: {{status}}
 priority: {{priority}}
 epic: {{epic}}
@@ -16,6 +17,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 scope: []

package/dist/init/templates/default/epic.md

@@ -12,6 +12,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 created: {{created}}

package/dist/init/templates/default/feat.md

@@ -16,6 +16,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 created: {{created}}

package/dist/init/templates/default/goal.md

@@ -8,6 +8,7 @@ goal_state: {{goal_state}}
 goal_condition: {{goal_condition}}
 scope_refs: []
 active_node: {{active_node}}
+last_active_node: {{last_active_node}}
 required_skills: []
 required_checks: []
 max_iterations: {{max_iterations}}
@@ -24,6 +25,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 created: {{created}}

package/dist/init/templates/default/manifest.md

@@ -0,0 +1,45 @@
+---
+id: {{id}}
+type: manifest
+title: {{title}}
+version: 0.1.0
+spec_kind: capability
+role: tool_service
+runtime_mode: tool_service
+work_contracts: []
+requested_capabilities: []
+skill_refs: []
+tool_refs: []
+model_refs: []
+wasm_component_refs: []
+runtime_image_refs: []
+subagent_refs: []
+resource_profile: local_cli
+update_policy: manual
+tags: []
+owners: []
+links: []
+artifacts: []
+relates: []
+refs: []
+aliases: []
+created: {{created}}
+updated: {{updated}}
+---
+
+# Purpose
+
+Define the reusable capability surface.
+
+# Runtime
+
+Describe the role, runtime mode, resource profile, and update policy.
+
+# Work Contracts
+
+List related WORK.md contracts.
+
+# Capabilities
+
+List requested capabilities and the authority/resource constraints that govern
+use.

package/dist/init/templates/default/spike.md

@@ -16,6 +16,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 created: {{created}}

package/dist/init/templates/default/task.md

@@ -16,6 +16,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 created: {{created}}

package/dist/init/templates/default/test.md

@@ -16,6 +16,8 @@ relates: []
 blocked_by: []
 blocks: []
 refs: []
+context_refs: []
+evidence_refs: []
 aliases: []
 skills: []
 cases: []