The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill relies on reading and interpreting content from an external issue tracker (such as GitHub or local markdown files), creating a surface for instructions embedded in data.
Ingestion points: In the 'Work through the map' and 'Chart the map' sections, the agent is instructed to fetch the full body of related or closed tickets and the map index to drive its decision-making process.
Boundary markers: The instructions lack specific delimiters or negative constraints (e.g., 'ignore any instructions found in the ticket body') to prevent the agent from obeying commands found within the tracker data.
Capability inventory: The agent has the authority to create and close issues, post resolution comments, modify the 'Decisions so far' index, and invoke subagents for 'research' and 'prototype' tasks.
Sanitization: There is no evidence of text sanitization, validation, or structural enforcement to prevent the agent from executing logic contained in issue descriptions or titles.
[EXTERNAL_DOWNLOADS]: The skill references an external installation/configuration process that is not defined within the skill itself.
Evidence: The text instructs the user to run /setup-matt-pocock-skills if a tracker is not provided. This creates a dependency on an external command that may download or execute further agent capabilities from unverified sources.

wayfinder