The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied bug descriptions and error symptoms to guide a multi-phase debugging and fixing process. It also incorporates historical context from project "memory" files. Because the skill has the capability to modify the filesystem and execute shell commands, maliciously crafted input could potentially influence the agent to perform unintended actions.
Ingestion points: The skill captures $ARGUMENTS as the symptom statement in SKILL.md and reads files from the docs/memory/ directory via the docs-index.sh utility.
Boundary markers: The instructions do not specify the use of delimiters or explicit boundary markers when interpolating user-supplied symptoms into the agent's context.
Capability inventory: The skill utilizes a wide range of powerful tools including Write, Edit, and Bash (allowing for package management, compilation, and test execution across multiple languages).
Sanitization: No explicit sanitization, validation, or filtering of the symptom input or memory file content is defined in the instructions.

systematic-debugging