The Agent Skills Directory

[METADATA_POISONING]: The YAML frontmatter in SKILL.md attributes the skill to https://github.com/Jeffallan, which contradicts the provided author context of alismx.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process untrusted external data such as error messages, stack traces, and log entries, creating an attack surface where malicious instructions could be embedded in the data being analyzed.
Ingestion points: SKILL.md and references/systematic-debugging.md instruct the agent to gather and read complete error messages and logs from the environment.
Boundary markers: The instructions do not specify any delimiters or safety prompts to ensure the agent ignores potential instructions contained within the logs or error messages.
Capability inventory: The skill encourages the use of powerful command-line tools including pdb, node --inspect, dlv, and git, which could be abused if the agent follows instructions hidden in processed data.
Sanitization: There are no instructions for sanitizing or filtering external content before the agent processes it.

debugging-wizard