Replace literal `.claude/skills/<owner>/...` paths in SKILL.md with the
`${CLAUDE_SKILL_DIR}` variable that Claude Code substitutes at invocation
time. Same-skill references become `${CLAUDE_SKILL_DIR}/<rest>`,
cross-skill references (erf-* → epf-*) become
`${CLAUDE_SKILL_DIR}/../<other>/<rest>`. All paths now wrapped in double
quotes to handle install locations with spaces.
This unblocks plugin-mode installation: literal `.claude/skills/...`
paths fail when the skill is loaded from `~/.claude/plugins/cache/...`
or from a personal `~/.claude/skills/` install. Drop-in mode continues
to work because Claude Code resolves the variable in all install scopes.
Verified via pilot:
- Project drop-in (cc-1c-skills repo)
- Personal `~/.claude/skills/cf-info/`
- Plugin via `/plugin marketplace add <local-path>`
Scope: 61 SKILL.md, 125 path replacements (8 cross-skill).
Scripts unchanged — they already use \$PSScriptRoot and ../<sibling>
patterns that resolve correctly under the bundled cache layout.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The intro paragraph on line 14 duplicated the frontmatter description
field. Per Anthropic best practices, the body should contain only
implementation details — the heading and frontmatter are sufficient.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Drop "см. /erf-*" from epf-build/epf-dump and "Использует общий скрипт
из /epf-*" from erf-build/erf-dump to prevent the model from reading
duplicate skill files and wasting context.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add "Используй когда..." triggers to epf-build, epf-dump, erf-build, erf-dump
- Remove "Коды возврата" sections (trivial info Claude already knows)
- Remove Hierarchical tree diagrams from dump skills (Claude sees output directly)
- Fix Windows backslashes to forward slashes in examples
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dump in an empty database irreversibly loses reference types (CatalogRef,
DocumentRef, etc.) — they get converted to xs:string. Instead of silently
creating an empty temp database, the script now exits with an error
explaining that a real database is required.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add stub-db-create script (.ps1/.py) that scans EPF XML sources for
reference types (CatalogRef, DocumentRef, EnumRef, etc.) and generates
a minimal 1C configuration with metadata stubs. Supports 14 metadata
types including registers, charts, defined types.
epf-build/erf-build: if no database specified, auto-create stub DB
with matching metadata, build EPF, then cleanup temp DB.
epf-dump/erf-dump: if no database specified, create empty DB with
warning that reference types will be converted to strings.
SKILL.md updated: prefer real database from .v8-project.json first,
fall back to auto-created stub only when unavailable.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Bash on Windows strips backslashes in unquoted paths, causing
"file not found" errors. Forward slashes work cross-platform.
Also fix pwsh → powershell.exe in epf-init and epf-add-form.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Wrap all remaining 1cv8.exe CLI calls with Start-Process .ps1 scripts
to fix Cyrillic parameter corruption in PowerShell 5.1 when using & operator.
New scripts (7):
- db-update.ps1: DESIGNER /UpdateDBCfg with Dynamic/Server/WarningsAsErrors
- db-dump-cf.ps1: DESIGNER /DumpCfg with Extension support
- db-load-cf.ps1: DESIGNER /LoadCfg with Extension support
- db-create.ps1: CREATEINFOBASE with UseTemplate/AddToList
- db-run.ps1: ENTERPRISE with Execute/CParam/URL (background, no wait)
- epf-build.ps1: DESIGNER /LoadExternalDataProcessorOrReportFromFiles (shared with erf-build)
- epf-dump.ps1: DESIGNER /DumpExternalDataProcessorOrReportToFiles (shared with erf-dump)
Updated SKILL.md (9): db-update, db-dump-cf, db-load-cf, db-create,
db-run, epf-build, epf-dump, erf-build, erf-dump — all now reference
powershell.exe -NoProfile -File invocation with parameter tables.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add priority step for user-provided connection parameters (path, server)
before name-based lookup in .v8-project.json. After task completion,
offer to register untracked databases via /db-list add.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace duplicated "Разрешение базы данных" sections in 8 db-* skills
with a compact "Параметры подключения" block referencing .v8-project.json.
Add Git branch-based database resolution priority (databases[].branches
matched against current branch before falling back to default).
Migrate epf-build, epf-dump, erf-build, erf-dump from V8_PATH/V8_BASE
environment variables to .v8-project.json — registered databases are now
preferred over empty build bases, solving XDTO errors with reference types.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Nick ShirokovandClaude Opus 4.7