Section 10.4 said nothing about inclusion being textual and repeated, so an
author met the behaviour only by reading rendered output carefully. That
silence was the defect.
Writing it up got the mechanism wrong at first. I documented a dependency
"reached by two paths" through two include chains, and wrote a detector that
counted composer calls to match. Tested it against the real case that motivated
the task — helix/repo-advance — and it did not fire, though the conventions
block still rendered twice.
The actual mechanism is subtler. helix/commit-sync declares its own
`conventions` input; when composed, resolution passes the including package's
already-resolved values down by name, so it *inherits* the outer text rather
than resolving its own include, and renders it again. Nothing was included
twice; the text appeared twice regardless.
Both mechanisms are now documented with worked diagrams. duplicate_inclusions
detects both — included values that are equal, and an included value contained
within another — and resolve, render and eval warn.
Nothing is deduplicated, as leaned. Deduplicating means choosing which
occurrence survives, since position in a prompt carries meaning, and deciding
what happens when two paths select different versions of the same dependency,
which section 10.3 permits. That is resolver behaviour and section 10.4 keeps
composition declarative.
Verified by reintroducing the diamond in a scratch copy of repo-advance, which
warns, and confirming the shipped factored collection stays silent.
Tests 90 -> 95.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
CANP-WP-0004. The operator asked canned-prompts to build a database of
versioned prompts recording where each came from and when — the first step
toward a platform for collaborative prompting.
The format had nowhere to put that. `provenance` records who wrote a prompt and
where the idea came from; nothing recorded how a copy arrived in a particular
store. Section 20.3 now specifies an `index.yaml` as store metadata rather than
package data: how a copy arrived differs for every consumer, and recording an
arrival must never rewrite the package that arrived.
`add`, `install` and `publish` record registry, id, version, name, source,
method, first-inclusion date, and the package's declared author, source and
licence — the last three copied so a listing is readable without opening every
package. A new `index` verb lists it. `included_at` is never overwritten; a
re-run updates `last_seen_at`, because when a package first entered a
collection is a fact about history rather than about the last command run.
Tests 84 -> 90.
Also records two findings from actually using the format:
CANP-WP-0004-T03 — inclusion has no deduplication, so a diamond dependency
renders shared content once per path. Found by composing a real collection.
Not fixed here: deduplicating means choosing which occurrence survives and
deciding what happens when two paths resolve different versions, which is
resolver behaviour that section 10.4 deliberately avoids. Handed to
CANP-WP-0005 with a leaning: document it, warn at validation time, do not
deduplicate.
The add_ons workaround in practice/pqrst-estimate is evidence about section
23's deferred "richer template syntax" — an optional appendix has to be an
input with an empty default, because CPF has no conditionals.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Found while verifying the repo against INTENT.md's eight success criteria for
a first practical release, after CANP-WP-0002 finished.
Criterion 5 is "publish a version to a registry and install it elsewhere".
Installing practice/pqrst-estimate into a fresh catalog reported success, and
rendering it then failed with "included package not found:
practice/house-style". CANP-WP-0002-T03 introduced this — composition made a
package installable but unrenderable, and nothing said so at install time.
The error was honest, but reaching it after a successful-looking install is
the silently-incomplete failure section 23 now names as the thing this format
avoids.
`add` and `install` now name declared prompt dependencies the target catalog
cannot satisfy, including a package present at no matching version. They do
not fetch anything: section 10 leaves dependency resolution to the consumer,
and auto-installing transitively is a resolver — a separate decision from
refusing to hand over a package that looks fine and is not.
Whether `install` should offer `--with-dependencies` is left open as a design
question rather than smuggled in as a bug fix.
Tests 81 -> 84. Recorded as CANP-WP-ADHOC-2026-09-06.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Closes the workplan. The format becomes `canned-prompt/v0.2`, and packages
declaring v0.1 remain valid — everything added across T01-T05 is additive, so
a v0.1 package means exactly what it always meant. That is the MINOR case
section 17 itself describes.
The spec file loses its version suffix: CannedPromptFormat-v0.1.md becomes
CannedPromptFormat.md, with the revision stated inside. One stable path that
never breaks a link, and no rename per revision; the version belongs in the
`format` string where tools actually read it.
Section 23 is 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. "Decided against" holds template inheritance
alone, because calling it deferred would misdescribe it — reopening it means
overturning a decision and answering four recorded objections, not filling a
gap.
Section 23 also names the two habits the five decisions turned out to 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.
The eval-rubric and registry-manifest schemas keep their own v0.1. They are
new in this revision and sit on their own version lines.
Reference CLI: ACCEPTED_FORMATS; an unknown revision is rejected naming what is
accepted. Tests 78 -> 81.
Example packages declare v0.2 and are bumped 0.1.0 -> 0.1.1 and 0.2.0 -> 0.2.1
as section 17 PATCH — metadata corrections with behavior unchanged.
Also refreshes section 22's worked example, which 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
composition illustration.
CANP-WP-0002 is finished. CANP-WP-0003 carries forward the one residual: the
default registry's basename-derived name reads as `registry:`.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Two defects from the original review of the seed.
Prerelease ordering was worse than first recorded. parse_semver returned
(major, minor, patch, raw_string), so 1.0.0-rc1 and 1.0.0 tied on the numeric
fields and then compared as strings — "1.0.0-rc1" > "1.0.0". A release
candidate therefore shadowed its own release for `newest` and for `>=`, not
just for the no-version case.
parse_semver now returns a SemVer section 11 precedence key: numeric fields, a
release/prerelease rank, then dot-separated prerelease identifiers with
numeric ones compared numerically. Build metadata is ignored. Beyond ordering,
prereleases are excluded from `any`, `newest` and `>=` entirely; only an exact
pin selects one, so publishing a release candidate never changes what existing
consumers resolve to. A package holding only prereleases now says so rather
than reporting a bare not-found.
Packaging copied the whole source directory, so a stray .git, virtualenv or
scratch file landed in the catalog and registry. Section 2 already required
otherwise — tools MUST ignore unknown non-reserved files unless a manifest
field references them — so this is conformance rather than a new rule. What is
new is that omissions are reported instead of silent:
not packaged (not a reserved path, not referenced by the manifest): .git/, .venv/, notes.txt
LICENSE joins the reserved paths. Strict packaging would otherwise drop a
package's license text while faithfully copying its `license` field, which
contradicts section 14's instruction to surface licensing on publish and
install.
Spec: 2 (LICENSE, packaging obligation, reporting), 17.1 new, 10.3 note.
Reference CLI: parse_semver rewritten with is_prerelease; select_version and
pick_version updated; copy_package and report_skipped replace copy_immutable.
Tests 65 -> 78.
examples/pqrst-estimate carries a LICENSE and a license field, exercising the
new reserved path.
Also fixes a leaked loop variable in package_members that would have reported
a bad `template` path as an `evals` error.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
The answer to the capability question was conditional: keep both fields if
they carry the required/observed distinction, fix the terminology if they do
not. They did not.
Section 9 opened with "records known requirements or observations", mixing
both in one field — `models` was observational ("known to be compatible or
evaluated") while `capabilities` was prescriptive ("expected from the
execution environment"). Section 10 then described dependencies as what a
prompt "expects". Both fields said expected, so the overlap was real
ambiguity rather than redundancy, and the fix is terminology.
`dependencies` now means **required**; `compatibility` means **observed**. A
consumer must not refuse to run a package because its environment is absent
from a compatibility list. The same capability name may legitimately appear in
both: required to run at all, and separately observed to work well on
particular models. `compatibility.aliases` records the same capability under
other names, so a consumer can recognize a requirement its environment labels
differently.
Dependencies now have three kinds, separated by what the format can do about
them: `prompts` it resolves by id and version; `context` names what it does
not package at all; `capabilities` are what the environment must be able to
do. Context entries use `name` rather than `id`, because nothing can look them
up, and `description` is required because nothing else can explain an
unpackaged dependency. A capability takes no version and no
`requirement: generate` — it is not an artifact and cannot be fetched, pinned
or generated. Capability names are free-form kebab-case, validated for shape
and not membership, exactly as tags are.
Section 10.1 also draws the line the format had never stated: an input is
content the caller passes for one use; a context dependency is a standing fact
about the environment.
Spec: 9 rewritten, 9.1 and 10.1 and 10.2 new, 10 reframed, 18 (rules 19-20),
4 updated. Former 10.1/10.2 renumbered to 10.3/10.4 with cross-references.
Reference CLI: validate_capabilities, validate_context_dependencies, and a
`resolve` section listing required capabilities and context under "this tool
cannot verify these" rather than implying it checked. Tests 51 -> 65.
Also drops an invented `session-review` capability from the example package in
favour of an honest `long-context` observation.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
`evals/` was a reserved path holding unvalidated blobs: section 12 named the
directory and gave an illustrative snippet, but nothing was specified, so no
tool could act on an eval file.
Every eval file must now declare a `schema`, and CPF defines exactly one —
`canned-prompts/eval-rubric/v0.1`. Unrecognized schemas stay legal and are
skipped rather than rejected, so the format gains something actionable without
becoming an evaluation language, which remains a non-goal.
The schema splits along the same seam as T01 and T03. Render checks
(`contains`, `not_contains`, `resolves_all`) assert properties of the rendered
prompt text, need no model, and are therefore run by the reference CLI. Output
criteria describe a good result and are declared but not run, because judging
them requires a model. That division is now the format's consistent answer to
"deterministic locally, or not".
An eval references a fixture already declared in the manifest's `examples`
rather than carrying its own copy, so an example that is also an eval fixture
stays honest — both break together.
An eval declares assessment and must not record outcomes. Results are run
evidence and live outside the immutable package, per INTENT.md and section 17.
Spec: 12 rewritten with 12.1, 18 (rules 17-18), 21 (`eval` verb).
Reference CLI: read_eval, validate_eval, load_example_values,
run_render_checks, cmd_eval; a failed render check exits non-zero.
Tests 42 -> 51.
examples/pqrst-estimate/evals/quality.yaml is a real eval with four render
checks and four output criteria, and it passes.
Fixes a latent bug reaching a fixture exposed: coerce_value assumed every
value was a command-line string, so a YAML fixture carrying a real type
(include_rationale: true) crashed on .lower(). Typed values are now validated
but not re-parsed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
T01 had already delivered half of composition without naming it: a derived
default binds an input to a prompt dependency, which is transclusion — run
package B, use its output. What was missing was the deterministic half.
`include` inlines another package's rendered template as text. No model is
involved, so the reference CLI can actually perform it, and a shared preamble,
rubric or style block becomes a versioned package instead of copied text. This
is the concrete way to honor INTENT principle 9 without any runtime. `derive`
stays as it was. Both are input defaults, so composition reuses the resolution
machinery rather than adding a second one.
No template inheritance. Four of this repo's own documents argue against it:
INTENT principle 3 (hidden context defeats reuse), section 19's "make package
contents visible before execution", section 17's requirement that behavior
changes produce a new version, and the non-goal on range resolution.
Version selectors: an exact pin is the expected form, with `any`, `newest` and
`>= X.Y.Z` as explicit opt-ins so looseness is written rather than implied by
absence. Selectors are evaluated per dependency against what is available —
no solver, no cross-dependency constraint satisfaction — which is what keeps
them outside the range-resolution non-goal, and the spec says so.
Also defines `type` (template | fragment), which appeared once in the section 4
manifest surface and was specified nowhere.
Spec: 3.2 (type), 5.1 (inclusion resolution rule, renumbered), 6.1 (included
default), 10.1 and 10.2 (new), 18 (rules 14-16), 21.
Reference CLI: validate_version_selector, select_version,
prompt_dependencies replacing prompt_dependency_ids,
check_composition_reference, CatalogComposer with cycle detection, and
resolve_inputs gaining composer= and inherited=. Tests 21 -> 42.
Examples: house-style is a real fragment package; pqrst-estimate composes it
and is bumped 0.1.0 -> 0.2.0 per section 17.
Fixes an ordering bug found while testing: inputs resolved before parameters,
so an included package could not see the including package's parameters and
silently fell back to its own defaults — the fragment rendered tone=neutral
where the including package said blunt. Parameters now resolve first; the
report still lists inputs first.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Section 17 already defined what a registry does when the same id@version is
republished; nothing defined what a *consumer* does. That is where namespace
conflict actually bites — in the catalog, after installing from two
registries.
Identity is now registry-scoped: an id names a package within a registry, the
way a path names a file within a repository. A local-first format with no
signing, no federation and no central authority cannot enforce global
uniqueness, and an unenforceable guarantee is worse than none — it invites
consumers to conflate two packages that merely share a name. A qualified
`<registry>:<id>` reference distinguishes them, and `:` is now barred from ids
so the separator stays available.
Installing the same id from two registries is therefore not a conflict. The
catalog is namespaced by registry and keeps both.
Ownership is registry policy, not package data. An optional `registry.yaml`
names a registry and records namespace claims. Those claims are explicitly
descriptive — a filesystem registry cannot authenticate a publisher, and
`publish` says so rather than implying it checked. Keeping the claim out of
packages leaves artifacts free of unverifiable assertions of authority, and
means package semantics do not change when a hosted registry appears later.
Spec: 3.2 (registry-scoped identity, qualified references), 17 (immutability
scoped to a registry), 20.1 and 20.2 (new), 18 (registry-manifest
validation), 21 (qualified references, reserved `local` name).
Reference CLI: parse_reference, check_registry_name, read_registry_manifest,
registry_name, namespace_policy; registry_package_path and
catalog_package_path split; resolve_installed reports ambiguity and returns
the source registry; iter_catalog; `add --as`; closed-namespace warning on
publish. Tests 11 -> 21.
The catalog layout changed. An existing catalog is detected and reported with
instructions rather than failing as "package not found". Signing, trust
scoring and federation remain non-goals and were not touched.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Closes the gap that made optional inputs unusable: rendering rule 5.1(4)
made any unresolved placeholder an error while inputs had no `default`, so
an input marked `required: false` and referenced from the template failed
every render in which the caller omitted it — including the spec's own
section 4 example.
Section 10 already carried the derivation mechanism (`requirement: generate`,
resolution deliberately undefined), so a derived default needed a binding
rather than a new concept: the input's default names a declared prompt
dependency.
Spec:
- 5.1 rewritten as "Resolution and rendering". Resolution may be
non-deterministic and must report what it derived; rendering is
deterministic and must not derive. A tool that handles only supplied
values and static defaults is stated to be conforming.
- 6.1 (new) covers both declaration forms. Reference form is preferred, with
the reason stated — an inline prompt is anonymous, so unversioned,
unprovenanced and un-evaluable — and validators should warn when a
published package derives inline.
- A derived default may declare a static fallback `value`. Without one the
input stays unresolved, which is an error; derivation never silently
yields empty content.
- 4, 10, 18 (rules 11-13), 19 (two new MUST NOTs), 21 updated accordingly.
Reference CLI:
- New `resolve` verb reporting the origin of every value.
- `Resolution` dataclass and `resolve_inputs`; `resolve_values` kept as a
wrapper so existing callers are unaffected.
- `render` refuses with a specific error naming underivable inputs rather
than substituting empty text.
- Tests 3 -> 11. Example package lifecycle re-verified end to end.
INTENT.md is unchanged: splitting resolve from render preserves success
criterion 4 (deterministic rendering) as written.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
- .repo-classification.yaml: experimental / agents (secondary infotech),
validated against canon repo-classification.allowed.yaml v1.1. Clears
consistency checks C-24 and C-35.
- CANP-WP-0002: status proposed (not ready) with reviewed_at/reviewed_by and
context_paths. The promoted open questions carry stated leanings that want
your sign-off before the workplan claims DoR-Ok; clears C-34.
- WORK-RECORDS.md and workplan state_hub ids as written by fix-consistency.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
Register canned-prompts under agents / practice (topic
c1d199b6-55ee-4db6-b49e-257a9f0f15ac, workplan prefix CANP-WP) via
`statehub register`, then replace the generated placeholders with
repo-specific facts.
- SCOPE.md: real boundaries drawn from INTENT.md's deliberate boundary,
current state (spec v0.1 + reference CLI, 3/3 tests pass, example
round-trips), and the developer workflow.
- AGENTS.md: drop the unresolved {CREDENTIAL_ROUTING} template token left
by the generator.
- CANP-WP-0001: bootstrap tasks closed.
- CANP-WP-0002: new workplan carrying the five § 23 open questions promoted
from "experience will decide" to "decide for v0.2" — optional-input
defaults (static or derived), registry namespaces/ownership, prompt
composition, canonical eval schemas, typed context/dependency contracts —
plus two reference-implementation conformance defects found in review
(prerelease versions sort as newest; copy_immutable packages the whole
source directory).
Also lands the previously untracked seed: INTENT.md, the CPF v0.1 spec,
the reference CLI, and examples/pqrst-estimate.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
