The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data such as technical specs, bug investigations, and existing tracker issues to generate or reconcile tickets. This creates an attack surface where malicious instructions embedded in these documents could influence the agent's actions on the issue tracker.
Ingestion points: Technical briefs, specifications, bug investigation reports, and tracker issue content (identified in SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified for the processed inputs.
Capability inventory: The skill performs write operations (creating/updating tickets) and relationship management on external trackers like GitHub (identified in SKILL.md).
Sanitization: No sanitization or validation of the input content is described before it is used to influence tracker updates.
[DYNAMIC_EXECUTION]: The skill utilizes a dynamic discovery mechanism to locate and load companion procedures and libraries at runtime based on the project catalog, which represents a form of dynamic logic loading from computed paths.
Evidence: SKILL.md instructs the agent to discover companions by runtime/project catalog and read the maintained tracker procedure through the companion discovery gate.

write-tickets