Bsp logo

Bsp

Community
brake71
bsp

БСП / библиотечная интеграция 1С. In standard subsystems, exported common module ВызовСервера / СлужебныйПрограммныйИнтерфейс stability or client/server execution? Read fundamentals.md even without BSP name. Scope before opening: Explicit BSP API review, including method existence -> matching-version reference. A migration/update handler using a library to save without business logic/exchange registration -> read before write, even without a BSP name or version. Ошибка у поля формы и Отказ в решении со стандартными подсистемами -> read. Pure BSL, existing form element properties, platform-only exchange -> answer without opening this skill; platform-only exclusion applies only without BSP API review. References verify BSP 3.1.11; other versions need matching sources.

Overview

Publisherbrake71
Repository1c-ssl-skills
Skill namebsp
Stars
85
Forks
12
Bundled files
26
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 26 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by brake71 on GitHub. Read the source before you install it.

Installation

Install the Bsp AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/brake71/1c-ssl-skills/tree/main/skills/bsp
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/brake71/1c-ssl-skills.git /tmp/1c-ssl-skills
mkdir -p .claude/skills
cp -r /tmp/1c-ssl-skills/skills/bsp .claude/skills/bsp
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Bsp in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Bsp on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Bsp is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Using 1C:BSP 3.1.11 in application code

This router is ASCII-only so that its initial read is safe even with a legacy Windows stdout code page. The task references remain UTF-8 Russian, with exact BSL names, signatures, regions, parameters, examples and pitfalls.

Reading UTF-8 files on Windows

Use a native UTF-8 reader: rg -n . -- "PATH" reads a file; rg patterns and -A/-B/-C select a section. Replace PATH with the installed file's path; quote paths containing spaces. Alternatively, read with explicit Python UTF-8:

bash
python -X utf8 -c "import sys; from pathlib import Path; sys.stdout.reconfigure(encoding='utf-8'); print(Path(sys.argv[1]).read_text(encoding='utf-8'))" "PATH"

Get-Content -Encoding UTF8 controls input decoding, not captured stdout. Avoid PowerShell content cmdlets and piping native readers through Select-String, Select-Object or Where-Object: they may re-encode stdout. Filter inside rg or Python instead. Apply UTF-8 reading to project files too; do not combine a safe skill read with an unsafe README read. Do not change global shell settings.

Workflow

Scope precedence: explicit BSP API review, including checking whether a method exists, requires the selected reference even when the recommendation uses platform APIs. Apply the version boundary below first. Standalone platform-only requests without BSP API review are outside this workflow; answer them directly. The presence of BSL files or a form element alone is not an integration task. A library-based migration/update write is an integration task even without an explicit BSP name or version; route it to update.md after the version check.

  1. Check the requested BSP version. For a different version, explain that these 3.1.11 references cannot confirm its signatures and request matching documentation/export; this boundary check needs no API reference. For 3.1.11 or an unspecified version, select one primary reference from the task table. Add a second only for a genuinely cross-cutting task, such as printing in a background job.
  2. Bind the installed root to the parent of the absolute SKILL.md file you successfully opened. Build the selected path as INSTALLED_ROOT/references/REFERENCE.md; each skill has its own root. Open that file before an integration/API answer. Locate the scenario using native rg -n headings/context, then check its rules, signature, example and pitfalls. The router and remembered knowledge are insufficient. If a search fails, run the reference fallback below using that same absolute SKILL.md path. A project file listing may omit hidden .agents/; it does not prove the installed reference is absent. A system skill's path is not the root of this project skill. Declare the reference unavailable only after the fallback fails; otherwise use its scenario.
  3. If the user proposes a call or rule, check its original full Module.Method name and purpose. Identify mismatches explicitly. An API signature verified for 3.1.11 is not verification for another BSP version.
  4. Prefer the stable public programmatic-interface region. Internal or deprecated APIs require an explicit warning and no public alternative. Show overridable-module hooks as implementations: BSP calls them; application code implements them, not a direct qualified module call. Exact Russian region and suffix names are documented in fundamentals.md.
  5. If a configuration XML export containing CommonModules/ is available, additionally verify each recommended signature/region with bsp_api.py. Otherwise use the supplied reference and do not claim source verification.
  6. Answer in the user's language: API, client/server execution context, minimal BSL example and important constraints. Check each full BSP call name and argument order against the reference or available export. Take application-specific object/handler names from the task or project; do not invent symmetric BSP methods. Discuss absent/internal/forbidden calls in prose or inline code, not a runnable fenced BSL example.

The skill is self-contained and needs no developer documentation or source export. Resolve references/ and scripts/ relative to this installed SKILL.md, not the project's working directory. src/cf/ is only an example export path, not a consumer project requirement.

Reference fallback

Replace ABSOLUTE_SKILL_MD with the file you opened (use forward slashes), and REFERENCE.md with the selected table entry. This reads independently of cwd:

bash
python -X utf8 -c "from pathlib import Path; p=Path('ABSOLUTE_SKILL_MD').parent/'references'/'REFERENCE.md'; print(p.read_text(encoding='utf-8'))"

Task routing

Primary taskReference
Module suffixes, locating modules, subsystem map, stable API vs hooksreferences/fundamentals.md
User messages (including binding to a field during server validation), XML/JSON, attributes by reference, secure storage, strings, dates, temporary directoriesreferences/base-common.md; add references/forms-validation.md for additional-attribute validation
Background/long-running operations, progress, scheduled jobsreferences/longs-and-jobs.md
Object numbering prefix, infobase prefix, data areareferences/prefixes.md
Infobase upgrade/version, migration, update handlerreferences/update.md
Data exchange, exchange nodes, change registrationreferences/data-exchange.md
Electronic signatures, machine-readable powers of attorney (MCD), cryptography, DSS, signature checksreferences/esign-mcd.md
Contact information, addresses, OKTMO/KLADR, address classifierreferences/contact-info.md
Other classifiers, including countriesreferences/classifiers.md
Currencies, exchange rates, banks/accounts, work schedules/calendarsreferences/currencies-banks.md
External components, ODatareferences/external-components.md
Users, permissions, RLS, access groups/profilesreferences/users-access.md
Email, SMS, message templates, discussions/interactionsreferences/comms.md
Business processes and tasksreferences/bp-tasks.md
User-session termination, deleting marked objects, security profilesreferences/admin-tools.md
Infobase backupreferences/backup.md
Performance assessment, monitoring center, user activityreferences/perf-monitoring.md
Personal-data protection/destruction dates, classification labelsreferences/protection-pd.md
Connected commands, external reports/processorsreferences/commands-external.md
Printing, print manager, report variants, data composition (SKD)references/print-reports.md
Locking/unlocking form attributes, properties, change-prohibition datesreferences/forms-validation.md
Files, volumes, versioning, writing a filereferences/files-and-versions.md
Multilingual code, NStr, current language, localizationreferences/multilang.md
Duplicates, bulk modification, subordinate structure, reference replacementreferences/report-dedup.md

For an unmapped BSP task, start with fundamentals.md. It also lists intentionally uncovered subsystems. References are workflows, not just a searchable method catalog: use the complete scenario, not an isolated matching name.

Optional configuration-export verification

Replace SKILL_DIR with this installed skill directory, CF_EXPORT with the export root containing CommonModules/, and MODULE/METHOD with real names. The export is read-only and not bundled. Python UTF-8 mode keeps CLI output readable on Windows without changing global settings.

bash
# Signature, region, documentation, source path and line range.
python -X utf8 "SKILL_DIR/scripts/bsp_api.py" method METHOD --module MODULE --src "CF_EXPORT"

# One module's exported methods across all regions.
python -X utf8 "SKILL_DIR/scripts/bsp_api.py" module MODULE --src "CF_EXPORT"

# Modules exposing stable public API.
python -X utf8 "SKILL_DIR/scripts/bsp_api.py" modules --src "CF_EXPORT"

--module disambiguates same-named methods. --src is mandatory: no automatic export discovery; missing paths or CommonModules/ yield a nonzero exit code. module includes all regions; modules lists stable API modules only. The output marks hooks, internal and deprecated methods; retain these warnings.

A text search of subsystem XML can locate a common-module reference, but assigning Catalog/Document objects to a subsystem requires parsing its Item references, including nested subsystem XML, not a guessed grep match.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Bsp AI skill do?

БСП / библиотечная интеграция 1С. In standard subsystems, exported common module ВызовСервера / СлужебныйПрограммныйИнтерфейс stability or client/server execution? Read fundamentals.md even without BSP name. Scope before opening: Explicit BSP API review, including method existence -> matching-version reference. A migration/update handler using a library to save without business logic/exchange registration -> read before write, even without a BSP name or version. Ошибка у поля формы и Отказ в решении со стандартными подсистемами -> read. Pure BSL, existing form element properties, platform-o...

Why use Bsp on TypingMind?

Because you install it once and use it with any model. Bsp is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Bsp in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/brake71/1c-ssl-skills/tree/main/skills/bsp. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Bsp?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Bsp?

As many as you like. As long as a model supports skills, you can use Bsp with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Bsp AI skill free?

Yes. It is published on GitHub by brake71 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇