The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from an external findings file provided by the user via the issues_file argument (SKILL.md).
Ingestion points: The agent reads the entire issue file to establish a baseline and determine resolution paths for each finding.
Boundary markers: The instructions do not specify any delimiters or boundary markers to distinguish between data and potential instructions within the findings file.
Capability inventory: The skill has access to powerful tools including Bash, Write, and Edit, enabling it to modify the local filesystem and execute shell commands.
Sanitization: There is no mention of sanitizing or validating the contents of the issue file before the agent uses it to influence its coding and testing actions.
[DYNAMIC_EXECUTION]: The skill compiles and runs local code as part of its core loop (SKILL.md).
Execution method: It invokes cargo test and cargo clippy, which result in the compilation and execution of the Rust crate's source and test code.
Context: This execution is restricted to the local environment and is the intended primary purpose of the skill's repair workflow.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to run command-line interface operations (SKILL.md).
Commands: Specifically, it executes cargo test -p <crate> and cargo clippy -p <crate> -- -D warnings to verify fixes and ensure code quality.

bugfix-crate