diff --git a/CannedPromptFormat-v0.1.md b/CannedPromptFormat.md
similarity index 87%
rename from CannedPromptFormat-v0.1.md
rename to CannedPromptFormat.md
index 44e5f29..802e6b9 100644
--- a/CannedPromptFormat-v0.1.md
+++ b/CannedPromptFormat.md
@@ -1,8 +1,10 @@
-# Canned Prompt Format v0.1
+# Canned Prompt Format
 
-**Status:** Seed specification / experimental  
+**Revision:** v0.2 — packages declare `format: canned-prompt/v0.2`  
+**Status:** Experimental  
 **Project:** `canned-prompts`  
-**Purpose:** Portable packaging of reusable prompts and prompt templates.
+**Purpose:** Portable packaging of reusable prompts and prompt templates.  
+**Compatibility:** packages declaring `canned-prompt/v0.1` remain valid (§ 3.2).
 
 ## 1. Goals
 
@@ -18,7 +20,7 @@ A conforming package should be:
 - versionable;
 - extensible with examples, evals, dependencies, and provenance.
 
-CPF v0.1 specifies the **artifact format**. It intentionally does not specify model execution, agent orchestration, dependency resolution, registry transport, or evaluation engines.
+CPF specifies the **artifact format**. It intentionally does not specify model execution, agent orchestration, dependency resolution, registry transport, or evaluation engines.
 
 ## 2. Package layout
 
@@ -78,7 +80,7 @@ The canonical manifest is UTF-8 YAML named `prompt.yaml`.
 ### 3.1 Minimal manifest
 
 ```yaml
-format: canned-prompt/v0.1
+format: canned-prompt/v0.2
 id: review/code-review
 name: Code Review
 version: 1.0.0
@@ -99,13 +101,20 @@ Required fields are:
 
 #### `format`
 
-MUST be exactly:
+The current revision is:
 
 ```yaml
-format: canned-prompt/v0.1
+format: canned-prompt/v0.2
 ```
 
-for this specification.
+A package declaring `canned-prompt/v0.1` remains valid and MUST still be
+accepted. Everything v0.2 adds is additive — input defaults, composition, the
+eval schema, registry manifests — so a v0.1 package means exactly what it
+always meant. By § 17's own guidance this is a MINOR revision: new capability,
+unchanged contract.
+
+A consumer MUST reject a `format` it does not recognize rather than guessing at
+its meaning.
 
 #### `id`
 
@@ -125,7 +134,7 @@ engineering/architecture-review
 practice/pqrst-estimate
 ```
 
-Rules for v0.1:
+Rules:
 
 - lowercase ASCII is RECOMMENDED;
 - `/`, `-`, `_`, and `.` MAY be used;
@@ -138,7 +147,7 @@ Rules for v0.1:
 ##### Identity is registry-scoped
 
 An `id` names a package **within a registry**, the way a path names a file
-within a repository. CPF v0.1 does **not** claim that an id is globally unique:
+within a repository. CPF does **not** claim that an id is globally unique:
 `practice/pqrst-estimate` obtained from two different registries may be two
 different packages, and a consumer that draws on more than one registry MUST
 keep track of which registry each package came from.
@@ -200,10 +209,10 @@ rendered on its own; the field records the author's intent so that a consumer
 can warn when a package is used against it — rendering a fragment as a
 standalone prompt, or including a whole template where a fragment was meant.
 
-## 4. Complete v0.1 manifest surface
+## 4. Complete manifest surface
 
 ```yaml
-format: canned-prompt/v0.1
+format: canned-prompt/v0.2
 id: review/code-review
 name: Code Review
 version: 1.2.0
@@ -285,7 +294,7 @@ All fields other than the required fields in section 3.1 are optional.
 
 ## 5. Prompt template syntax
 
-CPF v0.1 uses deliberately small placeholder semantics:
+CPF uses deliberately small placeholder semantics:
 
 ```text
 {{ variable_name }}
@@ -337,7 +346,7 @@ always produce the same output. A consumer that satisfies a derived default
 
 **Rendering rules:**
 
-8. Values are substituted as text in v0.1.
+8. Values are substituted as text.
 9. A placeholder with no resolved value is an error.
 10. Template evaluation MUST NOT execute arbitrary code.
 11. Rendering MUST NOT derive values. A tool offering derivation MUST perform
@@ -350,7 +359,7 @@ static defaults only. Such a tool is conforming: it reports an input with an
 unsatisfied derived default and no static fallback as unresolved, which
 rule 9 makes an error.
 
-CPF v0.1 does not define conditionals, loops, filters, or functions. Implementations MAY offer richer rendering modes only when explicitly declared by an extension; they MUST NOT silently reinterpret a v0.1 template as executable code.
+CPF does not define conditionals, loops, filters, or functions. Implementations MAY offer richer rendering modes only when explicitly declared by an extension; they MUST NOT silently reinterpret a template as executable code.
 
 ## 6. Inputs
 
@@ -374,7 +383,7 @@ Fields:
 | `description` | no | Human-readable explanation |
 | `default` | no | Value used when the caller supplies none; see § 6.1 |
 
-Recommended v0.1 input types are:
+Recommended input types are:
 
 - `content`
 - `text`
@@ -382,7 +391,7 @@ Recommended v0.1 input types are:
 - `path`
 - `json`
 
-These are descriptive hints in v0.1. A runtime MAY use them for validation or adapters.
+These are descriptive hints. A runtime MAY use them for validation or adapters.
 
 ### 6.1 Input defaults
 
@@ -674,7 +683,7 @@ Recommended requirement values:
 - `optional`
 - `generate`
 
-`generate` means that a resolver MAY satisfy a missing dependency by invoking an appropriate generation process. **CPF v0.1 does not define how generation or dependency resolution works.**
+`generate` means that a resolver MAY satisfy a missing dependency by invoking an appropriate generation process. **CPF does not define how generation or dependency resolution works.**
 
 ### 10.3 Naming and pinning a dependency
 
@@ -851,7 +860,7 @@ provenance:
       version: 1.1.0
 ```
 
-The field is descriptive in v0.1. Registries SHOULD preserve provenance when publishing or mirroring packages.
+The field is descriptive. Registries SHOULD preserve provenance when publishing or mirroring packages.
 
 ## 14. Licensing
 
@@ -923,11 +932,11 @@ This guidance is intentionally advisory because prompt behavior is probabilistic
 
 ## 18. Package validation
 
-A v0.1 validator SHOULD verify at least:
+A validator SHOULD verify at least:
 
 1. `prompt.yaml` exists and parses as YAML;
 2. required manifest fields exist;
-3. `format == canned-prompt/v0.1`;
+3. `format` names a recognized revision (§ 3.2);
 4. `id` is non-empty and contains no path traversal;
 5. `version` is non-empty;
 6. `template` resolves to a regular file inside the package;
@@ -991,7 +1000,7 @@ Implementations SHOULD:
 
 ## 20. Registry model
 
-CPF v0.1 does not mandate registry transport.
+CPF does not mandate registry transport.
 
 A valid registry may be:
 
@@ -1098,7 +1107,7 @@ registry, an id is unambiguous by definition.
 
 ## 21. Reference CLI semantics
 
-The v0.1 reference tool uses two stores:
+The reference tool uses two stores:
 
 - **catalog** — packages locally available for search/show/render;
 - **registry** — packages available for publish/install.
@@ -1143,17 +1152,32 @@ These semantics are illustrative, not mandatory for other implementations.
 
 ## 22. Worked example
 
+This is `examples/pqrst-estimate` in the `canned-prompts` repository, abridged.
+It composes a shared style fragment, so it also demonstrates § 10.4.
+
 `prompt.yaml`:
 
 ```yaml
-format: canned-prompt/v0.1
+format: canned-prompt/v0.2
 id: practice/pqrst-estimate
 name: PQRST Estimate
-version: 0.1.0
+version: 0.2.1
 summary: Estimate how session effort was distributed across PQRST categories.
 template: prompt.md
 
+dependencies:
+  prompts:
+    - id: practice/house-style
+      version: ">= 0.1.0"
+      requirement: required
+
 inputs:
+  - name: house_style
+    type: content
+    required: false
+    default:
+      include: practice/house-style
+
   - name: session_summary
     type: content
     required: true
@@ -1166,12 +1190,22 @@ parameters:
 output:
   format: markdown
 
+compatibility:
+  capabilities: [long-context]
+
+license: MIT
+
+examples: [examples/basic.yaml]
+evals: [evals/quality.yaml]
+
 tags: [retrospective, agentic-coding, pqrst]
 ```
 
 `prompt.md`:
 
 ```markdown
+{{ house_style }}
+
 Review the following coding-session summary and estimate the distribution of
 session effort across PQRST. Percentages must sum to 100%.
 
@@ -1188,21 +1222,60 @@ Session:
 Include rationale: {{ include_rationale }}
 ```
 
-## 23. Open questions for v0.2+
+## 23. Open questions
 
-Experience should determine whether later revisions standardize:
+### Settled since v0.1
+
+`CANP-WP-0002` promoted five items from "experience will decide" to decisions.
+Each is now specified. This table records where the rule lives and what was
+chosen, so a later reader can find the reasoning and not just the rule.
+
+| Question | Decision | Where |
+|---|---|---|
+| Optional inputs were unusable | `default` on inputs, static or derived; resolution separated from rendering | § 5.1, § 6.1 |
+| Registry namespaces and ownership | Identity is registry-scoped; ownership is registry policy, never a package's own claim | § 3.2, § 20.1, § 20.2 |
+| Prompt composition and inheritance | Composition by reference in two kinds, `include` and `derive`; no inheritance | § 10.3, § 10.4 |
+| Canonical evaluation schemas | One schema, `canned-prompts/eval-rubric/v0.1`; render checks run, output criteria are declared | § 12.1 |
+| Typed context and dependency contracts | `dependencies` means required, `compatibility` means observed; `context` names what CPF cannot package | § 9.1, § 10.1, § 10.2 |
+
+Two habits recur across those decisions, and later revisions should follow them
+rather than rediscover them:
+
+**Separate the deterministic half from the rest.** `include` and `derive`,
+render checks and output criteria, resolve and render. Each pair splits at the
+same seam: what a tool can do with the package alone, and what needs a model. A
+consumer that cannot call a model is never second-class — it does the
+deterministic half completely and reports the rest honestly, rather than
+guessing or silently producing something incomplete.
+
+**A package never asserts what it cannot back.** Ownership belongs to the
+registry that admits a package, not to the package that would like to claim a
+namespace. `compatibility` records observation, never permission. A derived
+default is a request to a consumer, not an instruction the package executes.
+This is why § 19 can treat a package as content rather than as an actor.
+
+### Still deferred
+
+Experience should still determine whether later revisions standardize:
 
-- typed context/dependency contracts;
 - content macros;
-- prompt composition and inheritance;
-- registry namespaces and ownership;
-- cryptographic integrity/signing;
-- canonical evaluation schemas;
-- model capability vocabularies;
+- cryptographic integrity and signing;
+- model capability vocabularies — § 10.2 keeps capability names free-form,
+  validated for shape and not for membership;
 - run manifests and evidence formats;
 - deterministic compilation manifests;
-- trust/reputation signals;
+- trust and reputation signals;
 - federated discovery;
-- richer template syntax.
+- richer template syntax;
+- pattern-matching render checks — § 12.1 stops at `contains`, `not_contains`
+  and `resolves_all` rather than adding a matching language to a format whose
+  § 5 rests on not having one.
 
-Until practical usage forces these decisions, v0.1 should remain intentionally small.
+### Decided against
+
+Template inheritance (§ 10.4) is not deferred; it was considered and refused.
+Reopening it means overturning a decision, not filling a gap, and would have to
+answer the four objections recorded there.
+
+Until practical usage forces these questions, the format should remain
+intentionally small.
diff --git a/README.md b/README.md
index bc9cc1b..71bce90 100644
--- a/README.md
+++ b/README.md
@@ -5,7 +5,7 @@
 This bundle contains a first project seed for `canned-prompts`:
 
 - [`INTENT.md`](INTENT.md) — project mission, boundaries, principles, and success criteria.
-- [`CannedPromptFormat-v0.1.md`](CannedPromptFormat-v0.1.md) — experimental package-format specification.
+- [`CannedPromptFormat.md`](CannedPromptFormat.md) — the package-format specification, currently revision v0.2.
 - [`reference/`](reference/) — deliberately small Python CLI implementing the basic lifecycle.
 - [`examples/pqrst-estimate/`](examples/pqrst-estimate/) — a real package that can be used to exercise the implementation.
 - [`examples/house-style/`](examples/house-style/) — a `type: fragment` package that `pqrst-estimate` composes.
@@ -67,7 +67,7 @@ or command-level `--catalog` / `--registry` options.
 
 ## Resolution vs rendering
 
-The format separates the two steps (`CannedPromptFormat-v0.1.md` § 5.1).
+The format separates the two steps (`CannedPromptFormat.md` § 5.1).
 Resolution decides a value for every input and parameter and may be
 non-deterministic; rendering substitutes those values and always is. An input
 may declare a default that is either a static value or a *derived* one — a
@@ -165,7 +165,7 @@ requires (this tool cannot verify these):
 
 ## Registries and identity
 
-An id names a package *within a registry* (`CannedPromptFormat-v0.1.md`
+An id names a package *within a registry* (`CannedPromptFormat.md`
 § 3.2). The same id obtained from two registries may be two different
 packages, so the catalog keeps them apart and a bare id that matches more than
 one is reported as ambiguous rather than guessed. Qualify it when you need to:
diff --git a/SCOPE.md b/SCOPE.md
index 7e9503f..4443956 100644
--- a/SCOPE.md
+++ b/SCOPE.md
@@ -14,7 +14,7 @@ evidence, and is inspectable before use. See `INTENT.md`.
 
 ## In Scope
 
-- `CannedPromptFormat-v0.1.md` — the package-format specification.
+- `CannedPromptFormat.md` — the package-format specification (revision v0.2).
 - `reference/` — a deliberately small Python CLI (`add`, `search`, `show`,
   `render`, `install`, `publish`) over a filesystem catalog and registry.
 - `examples/` — real packages that exercise the format and the implementation.
@@ -36,12 +36,13 @@ Thin execution adapters are permitted; a general runtime is not.
 
 ## Current State
 
-- **Seed / experimental.** Spec v0.1 is written; the reference CLI implements the
-  full local lifecycle and its 3 tests pass. One example package
-  (`examples/pqrst-estimate`) round-trips add → search → show → render → publish
-  → install.
-- The reference implementation is spec-conformant on the points it covers; the
-  known open design questions are tracked in `workplans/CANP-WP-0002-*.md`.
+- **Experimental, revision v0.2.** The five open questions promoted from v0.1
+  are resolved and specified (`CANP-WP-0002`); what remains deferred is listed
+  in § 23. Packages declaring `canned-prompt/v0.1` remain valid.
+- The reference CLI implements the full local lifecycle across eight verbs and
+  its 81 tests pass. Two example packages (`practice/pqrst-estimate` and the
+  `practice/house-style` fragment it composes) round-trip add → search → show →
+  resolve → render → eval → publish → install.
 
 ## Developer Workflow
 
@@ -50,7 +51,7 @@ cd reference
 python3 -m venv .venv && . .venv/bin/activate
 pip install -r requirements-dev.txt
 
-python3 -m pytest -q                        # tests
+python3 -m pytest -q                        # 81 tests
 CANNED_PROMPTS_HOME=/tmp/cp \
   python3 canned_prompts.py add ../examples/pqrst-estimate   # smoke
 ```
@@ -61,6 +62,6 @@ There is no lint or build step yet; `pyproject.toml` declares the
 ## Getting Oriented
 
 - Intent and principles: `INTENT.md`
-- Format specification: `CannedPromptFormat-v0.1.md`
+- Format specification: `CannedPromptFormat.md`
 - Agent instructions: `AGENTS.md`
 - Workplans: `workplans/`
diff --git a/examples/house-style/README.md b/examples/house-style/README.md
index 3af72b2..ebf3360 100644
--- a/examples/house-style/README.md
+++ b/examples/house-style/README.md
@@ -21,7 +21,7 @@ inputs:
 
 Inclusion is deterministic — the fragment's template is rendered and inlined as
 text, with no model involved — so the reference CLI performs it during
-`resolve`. See `CannedPromptFormat-v0.1.md` § 10.4.
+`resolve`. See `CannedPromptFormat.md` § 10.4.
 
 Its `tone` parameter is inherited from the including package when that package
 declares one, and otherwise falls back to the default here.
diff --git a/examples/house-style/prompt.yaml b/examples/house-style/prompt.yaml
index a55d391..533e6e8 100644
--- a/examples/house-style/prompt.yaml
+++ b/examples/house-style/prompt.yaml
@@ -1,7 +1,7 @@
-format: canned-prompt/v0.1
+format: canned-prompt/v0.2
 id: practice/house-style
 name: House Style
-version: 0.1.0
+version: 0.1.1
 summary: >
   Shared style block for review and estimate prompts: evidence first,
   no filler, state uncertainty plainly.
diff --git a/examples/pqrst-estimate/prompt.yaml b/examples/pqrst-estimate/prompt.yaml
index 5e06966..9f08670 100644
--- a/examples/pqrst-estimate/prompt.yaml
+++ b/examples/pqrst-estimate/prompt.yaml
@@ -1,7 +1,7 @@
-format: canned-prompt/v0.1
+format: canned-prompt/v0.2
 id: practice/pqrst-estimate
 name: PQRST Estimate
-version: 0.2.0
+version: 0.2.1
 summary: 'Produce a post-session estimate of effort distributed across the PQRST categories
   for an agentic coding session.
 
diff --git a/reference/canned_prompts.py b/reference/canned_prompts.py
index 8bb0eb7..410e75b 100755
--- a/reference/canned_prompts.py
+++ b/reference/canned_prompts.py
@@ -19,7 +19,9 @@ from typing import Any, Iterable
 
 import yaml
 
-FORMAT = "canned-prompt/v0.1"
+FORMAT = "canned-prompt/v0.2"
+# v0.2 is additive, so a v0.1 package means exactly what it always meant (§ 3.2).
+ACCEPTED_FORMATS = ("canned-prompt/v0.2", "canned-prompt/v0.1")
 REGISTRY_FORMAT = "canned-prompt-registry/v0.1"
 LOCAL_REGISTRY = "local"
 REGISTRY_NAME_RE = re.compile(r"^[A-Za-z0-9][A-Za-z0-9._-]*$")
@@ -324,8 +326,11 @@ def validate_package(package_dir: Path) -> dict[str, Any]:
     if missing:
         raise CannedPromptError("missing required fields: " + ", ".join(missing))
 
-    if manifest["format"] != FORMAT:
-        raise CannedPromptError(f"unsupported format: {manifest['format']!r}")
+    if manifest["format"] not in ACCEPTED_FORMATS:
+        raise CannedPromptError(
+            f"unsupported format: {manifest['format']!r}; expected one of "
+            + ", ".join(ACCEPTED_FORMATS)
+        )
 
     package_id = manifest["id"]
     if not isinstance(package_id, str) or not package_id.strip():
diff --git a/reference/tests/test_canned_prompts.py b/reference/tests/test_canned_prompts.py
index ab0c67a..6a4a26d 100644
--- a/reference/tests/test_canned_prompts.py
+++ b/reference/tests/test_canned_prompts.py
@@ -762,3 +762,26 @@ def test_copy_refuses_to_overwrite(tmp_path: Path) -> None:
     cp.copy_package(src, dst, manifest, "package")
     with pytest.raises(cp.CannedPromptError, match="already exists"):
         cp.copy_package(src, dst, manifest, "package")
+
+
+# --- format revisions (§ 3.2) ---
+
+@pytest.mark.parametrize("declared", ["canned-prompt/v0.2", "canned-prompt/v0.1"])
+def test_both_format_revisions_are_accepted(tmp_path: Path, declared) -> None:
+    """v0.2 is additive, so a v0.1 package means exactly what it always meant."""
+    pkg = write_pkg(
+        tmp_path / "p",
+        BASE.replace("canned-prompt/v0.1", declared)
+        + "inputs:\n  - name: greeting\n    required: false\n    default: hi\n",
+    )
+    assert cp.validate_package(pkg)["format"] == declared
+
+
+def test_unknown_format_is_rejected(tmp_path: Path) -> None:
+    pkg = write_pkg(
+        tmp_path / "p",
+        BASE.replace("canned-prompt/v0.1", "canned-prompt/v9.9")
+        + "inputs:\n  - name: greeting\n    required: false\n    default: hi\n",
+    )
+    with pytest.raises(cp.CannedPromptError, match="unsupported format"):
+        cp.validate_package(pkg)
diff --git a/workplans/CANP-WP-0002-format-open-questions.md b/workplans/CANP-WP-0002-format-open-questions.md
index 85b6943..66208ad 100644
--- a/workplans/CANP-WP-0002-format-open-questions.md
+++ b/workplans/CANP-WP-0002-format-open-questions.md
@@ -4,7 +4,7 @@ type: workplan
 title: "Resolve CPF v0.1 open questions promoted for v0.2"
 domain: agents
 repo: canned-prompts
-status: proposed
+status: finished
 owner: codex
 topic_slug: practice
 created: "2026-09-06"
@@ -13,7 +13,7 @@ reviewed_at: "2026-09-06"
 reviewed_by: "claude"
 context_paths:
   - "INTENT.md"
-  - "CannedPromptFormat-v0.1.md"
+  - "CannedPromptFormat.md"
   - "reference/canned_prompts.py"
   - "examples/pqrst-estimate/"
 state_hub_workstream_id: "c87c8e27-8b11-5306-b687-86779ef3f1be"
@@ -21,7 +21,7 @@ state_hub_workstream_id: "c87c8e27-8b11-5306-b687-86779ef3f1be"
 
 # Resolve CPF v0.1 open questions promoted for v0.2
 
-`CannedPromptFormat-v0.1.md` § 23 currently lists twelve items as "experience
+`CannedPromptFormat.md` § 23 currently lists twelve items as "experience
 should determine". A review of the seed on 2026-09-06 (spec + reference CLI +
 `examples/pqrst-estimate`; 3/3 tests pass, full local lifecycle smokes clean)
 plus an operator interview promoted five of them from "wait and see" to
@@ -36,6 +36,11 @@ federated discovery, richer template syntax.
 format may *declare* a requirement; it must not specify the resolver, runtime,
 or execution engine that satisfies it.
 
+**Outcome.** All seven tasks are done. The format is now revision v0.2
+(`CannedPromptFormat.md`), the reference CLI covers eight verbs with 81 passing
+tests, and § 23 records what was settled, what stays deferred, and what was
+decided against. One residual is carried forward as `CANP-WP-0003`.
+
 ## Optional inputs need defaults
 
 ```task
@@ -373,16 +378,48 @@ what it genuinely has. Illustrations live in the spec; examples stay honest.
 
 ```task
 id: CANP-WP-0002-T06
-status: todo
+status: done
 priority: medium
 state_hub_task_id: "06032689-fc19-5fcd-a76e-5dbc87d02cd4"
 ```
 
-After T01–T05 land, replace § 23's flat twelve-item list with two sections:
-questions **being decided for v0.2** (each with its scoped question and stated
-leaning, referencing this workplan) and questions **still deferred** (the eight
-unpromoted items listed at the top of this file). Bump the spec status line if
-the format revision warrants it.
+**Decisions (operator, 2026-09-06):**
+
+- *Bump to `canned-prompt/v0.2`, accept both.* Everything added is additive, so
+  a v0.1 package means exactly what it always meant — the MINOR case § 17
+  itself describes. Tools accept both strings; new packages declare v0.2.
+- *Drop the version from the spec filename.* `CannedPromptFormat-v0.1.md` →
+  `CannedPromptFormat.md`, with the revision stated in the document. One stable
+  path that never breaks a link, and no rename per revision; the version lives
+  in the `format` string, where tools actually read it.
+
+Delivered:
+
+1. § 23 rewritten into three parts rather than the planned two. **Settled since
+   v0.1** tables the five resolved questions against where each rule now lives.
+   **Still deferred** carries the eight unpromoted items plus pattern-matching
+   render checks, deferred during T04. **Decided against** holds template
+   inheritance on its own, because "deferred" would misdescribe it: reopening
+   it means overturning a decision and answering four recorded objections, not
+   filling a gap.
+2. § 23 also names the two habits the five decisions share, so later revisions
+   follow them rather than rediscover them: *separate the deterministic half
+   from the rest*, and *a package never asserts what it cannot back*.
+3. Header now carries revision, status and a compatibility line; § 3.2 states
+   the acceptance rule and why v0.2 is MINOR; § 18 rule 3 checks for a
+   recognized revision rather than one literal string.
+4. Stale "CPF v0.1" phrasings throughout replaced with unversioned wording. The
+   `canned-prompts/eval-rubric/v0.1` and `canned-prompt-registry/v0.1` schemas
+   deliberately keep their own v0.1 — they are new in this revision and on
+   their own version lines.
+5. `reference/canned_prompts.py`: `ACCEPTED_FORMATS`; unknown revisions are
+   rejected with the accepted list. Tests 78 → 81.
+6. Example packages declare v0.2, bumped 0.1.0 → 0.1.1 and 0.2.0 → 0.2.1 as
+   § 17 PATCH — a metadata correction with intended behavior unchanged.
+
+**Also fixed.** § 22's worked example had drifted: it showed `pqrst-estimate`
+at 0.1.0 with no composition, contradicting the package actually in the repo.
+It now mirrors the real package and doubles as a § 10.4 illustration.
 
 ## Reference implementation conformance fixes
 
diff --git a/workplans/CANP-WP-0003-registry-naming-residual.md b/workplans/CANP-WP-0003-registry-naming-residual.md
new file mode 100644
index 0000000..4f99b17
--- /dev/null
+++ b/workplans/CANP-WP-0003-registry-naming-residual.md
@@ -0,0 +1,54 @@
+---
+id: CANP-WP-0003
+type: workplan
+title: "Default registry name reads as 'registry:'"
+domain: agents
+repo: canned-prompts
+status: proposed
+owner: codex
+topic_slug: practice
+created: "2026-09-06"
+updated: "2026-09-06"
+---
+
+# Default registry name reads as `registry:`
+
+Residual from `CANP-WP-0002` (`origin: residual`, `origin_ref: CANP-WP-0002`).
+
+## Give the default registry a better name
+
+```task
+id: CANP-WP-0003-T01
+status: todo
+priority: low
+```
+
+`CANP-WP-0002-T02` made a registry's name come from its `registry.yaml`, and
+otherwise from its directory basename (§ 20.1). The reference tool's default
+registry lives at `~/.canned-prompts/registry`, so its basename-derived name is
+`registry`, and every install from it reads:
+
+```text
+installed registry:practice/pqrst-estimate@0.1.0
+```
+
+That is honest — it is genuinely what the directory is called — but it reads
+poorly, and `registry:` is a confusing qualifier for the one registry most
+users will ever see.
+
+Deliberately not fixed in `CANP-WP-0002`: special-casing the default path to
+produce a prettier name would be exactly the kind of magic the § 20.1 rule was
+written to avoid, and it was not worth an exception mid-task.
+
+Options worth weighing:
+
+1. Write a starter `registry.yaml` when the default registry is first created,
+   naming it something like `default` or `local-registry`. Uses the documented
+   mechanism rather than bypassing it; the cost is a tool writing a file into
+   a store the user may consider theirs.
+2. Rename the default directory itself, so the basename is already good.
+   Changes an existing on-disk layout.
+3. Accept it and document it. Zero code, and the escape hatch — add a
+   `registry.yaml` — already exists and is documented.
+
+Option 3 is the current state. Pick deliberately rather than by default.