rlm-workflow
25SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A required reading step for work involving a specialised business or technical area.
Agent development framework for 1C
This repository also configures its own agents. See what 1c-agent-based-dev-framework tells them →
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A required reading step for work involving a specialised business or technical area.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A policy requiring a written specification before building features, changing architecture, or fixing complex bugs.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A rule to search the existing code and project before adding new code or a new query.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A coding rule for handling secrets, access tokens, cryptography, and user permissions.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A central record of limits for how coding sub-agents recover from failed fixes, testable guesses, and reviews.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A rule for writing code comments that explain the reason behind a decision.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A learning rule for maintaining patterns learned while using skills. It requires updates before skill use and after iterations.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A startup protocol for subagents that requires reading skills and rules before beginning work.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A decision rule for resolving conflicts and failures by following the designated source of truth—the information treated as authoritative.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A rule for checking БСП mechanisms before adding custom logic. БСП is the standard subsystem library used in 1C applications.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A test-first coding rule known as TDD, or test-driven development. It follows the cycle Red, then Green: first make a test fail, then change the code until it passes.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A cleanup rule for tests that write to a database. Zero-residue cleanup means the test must remove the data it created.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A troubleshooting rule for failures in Vanessa, an automated testing tool. It checks the registration log, visual output, and technical log.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A run rule for Vanessa tests after changing feature or test files, or using MCP. Vanessa is an automated testing tool.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A set of writing rules for Vanessa feature files. Features are structured descriptions of application behaviour that automated tests can execute.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A verification rule for a Security Warning shown in the registration log. Visual verification means confirming the warning through the visible result as well.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A data-isolation rule for Vanessa tests, which are automated tests for data-processing code.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A set of location rules for Vanessa feature files, which describe automated tests in a readable format.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A testing skill for YaxUnit database-writing tests that uses transactions. A transaction groups database changes so they can be handled as one unit during a test.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A procedure for maintaining a role-specific context file, which records useful names, working details, and how to resume a task.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A short workflow for small fixes that affect one file and fewer than 20 lines, without adding metadata objects or making architecture decisions.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A learning-review skill for recording lessons after work takes at least two attempts, including an error, a fix, and a successful result. It stores useful and harmful patterns in a project learning file.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A skill for designing and reviewing public APIs in 1C, a business software platform widely used in Russian-speaking organisations. An API is the defined way that other software connects to a system.
SteelMorgan/1c-agent-based-dev-framework
Skill Claude CodeCodex
A skill for designing and debugging background jobs in 1C, a business software platform widely used in Russian-speaking organisations. Background jobs run work away from the immediate user interaction.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: