fix: v0.11.1 — pascal/camel as distinct styles, context/ mirror sync

- ids.py: pascal → PascalCase (BoldVale), camel → true camelCase
  (boldVale). Previously camel incorrectly generated PascalCase.
- settings.toml updated to word_style = "pascal"
- context/ dogfood mirror synced: resolve_entity(), MCP server
  partial-ID lookups, session-handover SKILL.md, INDEX.md expansion
- aibox.lock bumped to v0.17.12 / processkit v0.10.0
- Templates mirror: context/templates/processkit/v0.10.0/ added

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: claude

AGENTS.md

@@ -263,7 +263,7 @@ call — no restart needed.
 
 | Preference | Value |
 |---|---|
-| ID word-pair style | `camel` — e.g. `BoldVale` |
+| ID word-pair style | `pascal` — e.g. `BoldVale` (also: `camel` → `boldVale`, `kebab` → `bold-vale`) |
 | ID datetime prefix | enabled — format `YYYYMMDD_HHMM` embedded in ID |
 | ID slug | enabled — content-derived from entity title |
 | Example ID | `BACK-20260409_1449-BoldVale-fts5-full-text-search` |

CHANGELOG.md

@@ -5,6 +5,31 @@ Versions follow [Semantic Versioning](https://semver.org/).
 
 ---
 
+## [v0.11.1] — 2026-04-11
+
+### Fixed
+
+- **`ids.py`: `pascal` and `camel` are now distinct word styles** —
+  `pascal` produces PascalCase (`BoldVale`, every word capitalised);
+  `camel` now correctly produces true camelCase (`boldVale`, first word
+  lowercase). Previously `camel` incorrectly generated PascalCase.
+  `id-management/config/settings.toml` updated to `word_style = "pascal"`.
+
+### Changed
+
+- **`context/` dogfood mirror synced** — `resolve_entity()` and
+  partial-ID lookup now live in the installed `context/` copies of
+  `index.py` and all MCP servers, consistent with the `src/`
+  implementation shipped in v0.10.0.
+- **`session-handover` SKILL.md** — log-entry writing steps updated
+  with `generate_id` call, date-sharded path derivation, and word-pair
+  ID template.
+- **`context/skills/INDEX.md`** — expanded with skill package layout
+  section and `_lib/` note.
+- **`aibox.lock`** — updated to aibox v0.17.12, processkit v0.10.0.
+
+---
+
 ## [v0.11.0] — 2026-04-11
 
 ### Added

HANDOVER.md

@@ -1,10 +1,10 @@
 [aibox]
-cli_version = "0.17.11"
-synced_at = "2026-04-10T12:38:48.050448+00:00"
+cli_version = "0.17.12"
+synced_at = "2026-04-10T21:51:18.112494+00:00"
 
 [processkit]
 source = "https://github.com/projectious-work/processkit.git"
-version = "v0.8.0"
+version = "v0.10.0"
 src_path = "src"
-release_asset_sha256 = "dbd0639aac7f2ac49708e770415a4d49b85ac2838df534a6e85a98b18b75697b"
-installed_at = "2026-04-10T12:38:48.050448+00:00"
+release_asset_sha256 = "93394a9bd73243dfa7ec273b3169d8bb77a28763995ad9fc88e88c152d96b965"
+installed_at = "2026-04-10T21:51:18.112494+00:00"

^

@@ -0,0 +1,123 @@
+---
+apiVersion: processkit.projectious.work/v1
+kind: Discussion
+metadata:
+  id: DISC-20260411_0044-ClearKeel-context-structure-primitive-taxonomy
+  created: '2026-04-11T00:44:00+00:00'
+  updated: '2026-04-11T00:44:00+00:00'
+spec:
+  question: How should the context/ directory, primitive taxonomy, and Zettelkasten
+    linking be structured and clarified in processkit?
+  state: resolved
+  opened_at: '2026-04-11T00:44:00+00:00'
+  closed_at: '2026-04-11T00:44:00+00:00'
+  participants:
+  - ACTOR-claude
+  - ACTOR-owner
+  outcomes: []
+---
+
+## Driving questions
+
+1. Why do agents create free-form files instead of using skills/MCP tools?
+2. Where do team descriptions, work instructions, PRDs, and research
+   results belong?
+3. What is the difference between an Artifact, a Work Instruction, and
+   a Note?
+4. What directory structure should `context/` have, and why does
+   `discussions/` exist as a top-level directory?
+5. Should Note be a separate primitive or a subtype of Artifact?
+6. Are Artifact entities pointers or self-hosted documents?
+7. Is the Zettelkasten linking mechanism (qualified note-to-note links)
+   missing from the Note schema?
+
+## Resolutions
+
+### R1 — AGENTS.md bloat is an antipattern
+Keep AGENTS.md as complete as necessary but as lean as possible.
+It should contain only what an agent needs at session start that
+cannot be derived from the project structure or entities. Anything
+with its own lifecycle, owner, or update cadence belongs in an
+entity.
+
+### R2 — Team configuration belongs in Actor entity files
+Model, role (PM vs developer), and parallelism pattern for AI actors
+belong in their Actor entity files — not in AGENTS.md. Actor profiles
+have their own update cycle; duplicating this in AGENTS.md causes
+immediate drift.
+
+### R3 — Work instructions are Artifacts
+A project-specific operational document (asciinema recording guide,
+release procedure variant) is a `kind: document` Artifact. No new
+primitive is needed. For formal repeatable processes, use a Process
+entity in `context/processes/`.
+
+### R4 — Artifact schema should acknowledge self-hosted content
+The schema description ("processkit does not store artifact content
+itself; just the pointer") is inconsistent with actual usage (the
+PRD artifact stores its full content inline). The schema should
+acknowledge two valid patterns:
+- **Self-hosted**: Markdown body after frontmatter; `location` may
+  self-reference or be omitted.
+- **Pointer**: deliverable lives elsewhere; `location` is the URL,
+  path, or storage ID.
+
+### R5 — Note state machine justifies keeping it separate from Artifact
+Notes have lifecycle mechanics (state machine, `review_due`,
+`promotes_to`) that Artifacts lack. These mechanics are what make
+the Zettelkasten capture-and-promote workflow function. Merging Note
+into Artifact would require adding optional lifecycle fields to
+Artifact, changing its character for non-note uses. Note stays a
+separate primitive, but the schemas should cross-reference the
+distinction explicitly.
+
+### R6 — Notes are not ephemeral; the type taxonomy needs alignment
+In Luhmann/Ahrens, permanent notes are never discarded — they are
+the knowledge base. The current `type` enum (fleeting, insight,
+reference, question) approximates but does not cleanly map to the
+three canonical types (fleeting, literature, permanent). The mapping
+should be made explicit in the schema description.
+
+### R7 — Qualified links are missing from the Note schema
+The Zettelkasten insight-generation mechanism depends on qualified
+note-to-note links, not just tags. Tags are for categorical
+discovery; links with context are for reasoning. The Note schema
+needs a `links` field:
+
+```yaml
+links:
+  - target: NOTE-xxx-yyy
+    relation: elaborates  # elaborates | contradicts | supports |
+                          # is-example-of | see-also | refines |
+                          # sourced-from
+    context: "One sentence explaining why this connection matters."
+```
+
+The `context` sub-field is non-optional in Luhmann practice.
+
+### R8 — context/ directory structure: one dir per primitive kind
+The current scheme (`context/workitems/`, `context/discussions/`,
+`context/notes/`, etc.) is correct. Each top-level directory maps
+1:1 to a primitive `kind:`, making it machine-readable without
+convention lookup. The indexer eliminates the need for
+subtype-based subdirectories within a kind. This is the "class /
+object" pattern: top level = kind, files within = instances.
+
+### R9 — Discussion is a legitimate separate primitive
+Discussion models structured deliberation (question → exploration →
+DecisionRecord). It is not a WorkItem (not a unit of work), not a
+Note (multi-party, structured, produces outcomes), not an Artifact
+(no deliverable). The epistemic arc is: Note → Discussion →
+DecisionRecord → WorkItem.
+
+### R10 — context/owner/, context/actors/, context/roles/ need
+documenting
+These directories are defined in schemas but absent from the
+AGENTS.md layout table. They should be added.
+
+### R11 — MCP config paths are stale after skills reorganization
+The `.mcp.json` at project root still references
+`context/skills/<name>/mcp/server.py` (pre-reorganization paths).
+After the `e86ad06` reorganization commit, the correct paths are
+`context/skills/processkit/<name>/mcp/server.py` (and category
+subdirs for non-processkit skills). The config must be regenerated.

aibox.lock

@@ -0,0 +1,15 @@
+---
+apiVersion: processkit.projectious.work/v1
+kind: LogEntry
+metadata:
+  id: LOG-loyal-vale
+  created: '2026-04-10T22:47:49+00:00'
+spec:
+  event_type: discussion.opened
+  timestamp: '2026-04-10T22:47:49+00:00'
+  summary: 'Opened Discussion ''DISC-bold-plum'': ''How should context/ directory
+    structure, primitive taxonomy (Artifact vs Note), Zettelkasten linking, and AGENTS.md
+    scope be clarified and improved in processkit?'''
+  subject: DISC-bold-plum
+  subject_kind: Discussion
+---