The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill implements a workflow where the agent generates BSL (1C:Enterprise) source code, writes it to the local filesystem as a CommonModule, and then loads it into the database for execution using the /db-load-xml and run_unit_tests tools.
[COMMAND_EXECUTION]: The agent is instructed to use the Bash tool to run external CLI utilities, specifically /db-load-xml and /db-update, to modify the 1C database configuration and apply code changes at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially contain malicious instructions or influence agent behavior.
Ingestion points: Test execution results (JSON/text) returned by the run_unit_tests tool in SKILL.md.
Boundary markers: None identified; output from tools is fed directly back into the agent's context without explicit delimiters or warnings.
Capability inventory: The skill possesses Write access to the filesystem, Bash access for shell commands, and arbitrary code execution within the 1C environment via mcp__1c-ai-debug__execute_1c_code.
Sanitization: No evidence of output sanitization or validation of the results received from the database tools before processing by the agent.

1c-test-runner