skill-creator-ext: Скилл для 1С BSL Framework

--- name: skill-creator-ext description: > 1C BSL Framework extension for skill-creator skill. Use together with the base skill-creator skill when creating or modifying framework skills (bsl-practices, tool-usage, spec-writing, agent-development-ext, etc.). Covers: framework skill categories, tool-usage skills replacing tool-registry, BSL content patterns, 1c-ai-agent-cli integration, project-specific skills. --- # Skill Creator — 1C BSL Framework Extension > **Base skill:** `skill-creator` (Anthropic). > Read the base skill first — it contains the general principles for creating skills. > This file adds **only** the 1C-specific flavor for the framework. --- ## 1. When to create a new skill | Trigger | Action | |---------|--------| | A new MCP tool appears and the agent does not know WHEN to use it | Create a `tool-usage/` skill | | A repeated anti-pattern in BSL code is found | Create a `bsl-practices/` skill | | The agent needs training on a specific configuration | Create a **project-specific** skill (inside the IDE project directory) | | A standard methodology is needed (SDD, TDD, review) | Create a `spec-writing/` skill or a new subfolder | | An external skill needs adaptation for the framework | Create a `*-ext/` extension | **Do NOT create** a skill if: - The information already exists in an existing skill — extend it instead - This is a one-off instruction — write a prompt, not a skill - This is a rule/policy (MUST/SHOULD) — create a rule, not a skill --- ## 2. Framework skill categories | Category | Directory | Purpose | Examples | |----------|-----------|---------|----------| | **BSL practices** | `skills/bsl-practices/` | Coding standards and patterns | `coding-standards`, `query-patterns` | | **Tool-usage** | `skills/tool-usage/` | When and how to use MCP tools | `syntax-checking`, `metadata-discovery`, `test-execution` | | **Spec-writing** | `skills/spec-writing/` | Documentation and specification standards | `spec-standard` | | **Extensions** | `skills/*-ext/` | Extensions of external skills (Anthropic, etc.) | `agent-development-ext`, `skill-creator-ext` | ### Extensions (-ext) — convention 1. The base skill is installed via `npx skills add` (updatable) 2. The extension lives at `<base-name>-ext/SKILL.md` inside `framework/skills/` 3. An extension **only adds to** the base skill; it does not duplicate its content 4. In the extension: `> Read the base skill first: <name>` --- ## 3. Skill creation process 1. **Analysis:** determine the category, check for duplication, identify dependencies 2. **Design:** name the file `kebab-case.md`, target 300 lines (max 500) 3. **Writing** — required sections: - YAML frontmatter - Title + 1–3 sentences describing the purpose - **When to apply** — trigger → action table - **Scenarios** — concrete examples - **Code samples** — right + wrong with explanation of WHY 4. **Integration:** place the file under `framework/skills/`, refresh skills in agents, run `python tools/install.py --list` --- ## 4. Tool-usage skills — replacing tool-registry Tool-usage skills are the **sole place** where MCP tools are described in the framework. ### Tool-usage skill structure ```markdown # [Title] — how to use MCP tools for [task] ## When to apply | Trigger | Action | |---------|--------| | The user asks to check syntax | Call `bsl.checkSyntax(uri)` | | The agent created/edited .bsl code | Run a syntax check automatically | ## MCP Tools | Tool | Purpose | Workarounds | |------|---------|-------------| | `bsl.checkSyntax` | Syntax checking | If the URI is missing, verify the encoding | ## Scenarios [Concrete call chains] ``` ### Principles 1. **Do not duplicate the MCP tool description** — it comes from MCP (`tools/list`) 2. **Focus on WHEN and WHY**, not parameter details 3. **Workarounds and pitfalls** are the main value 4. **Concrete scenarios** — call chains for common tasks --- ## 5. Skill formats | Type | Location | Frontmatter | Installation | |------|----------|-------------|-------------| | Framework skill | `framework/skills/` | `name`, `description`, `depends_on` | CLI via symlinks | | IDE skill | `.cursor/skills/` or `.agents/skills/` | `name`, `description` | `npx skills add` | | Extension (-ext) | `framework/skills/<name>-ext/` | `name: base-name-ext`, `description` | CLI, framework-managed | --- ## 6. Content patterns **Trigger → Action** (tool-usage): table `| Trigger | Action |` **Right / Wrong** (bsl-practices): two code blocks + explanation of WHY (what breaks if ignored) **Workflow with checklist** (methodologies): `- [ ] Step N: ...` **Conditional workflow** (branches): `Determine the type → branch A / branch B` --- ## 7. Project-specific skills Placement: `.cursor/skills/` or `.claude/skills/` at the project root. Anthropic SKILL.md format. **Include:** configuration architecture, local conventions, business rules, critical modules, integrations. **Do NOT include:** generic BSL standards (leave those in framework skills), secrets, information that ages quickly. --- ## 8. BSL specifics ### Platform limitations - BSL does not support inheritance — only common modules - There is no package manager — dependencies are wired through the configuration - Code is divided by contexts (server/client/external connection) - Metadata is declarative, code is imperative --- ## 9. New skill checklist - [ ] YAML frontmatter, name in kebab-case (ASCII) - [ ] Explains WHY, includes right/wrong examples - [ ] "When to apply" table (trigger → action) - [ ] No duplication with existing skills - [ ] Length ≤ 300 lines (max 500) - [ ] File under `framework/skills/`, agents updated, `install.py --list` shows the skill --- depends_on: [] ---