The Agent Skills Directory

[SAFE]: No high-severity malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill's core purpose is to restructure agent output into numbered lists and tables to reduce cognitive load.
[PROMPT_INJECTION]: The skill uses an indirect prompt injection surface by reading task data from a local file (AGENTS.md). If this file were modified by an untrusted source, the embedded content would be processed by the agent.
Ingestion points: The skill reads the Active Issues section from the project-level AGENTS.md file.
Boundary markers: The ledger data is enclosed within specific markers:  and .
Capability inventory: The skill allows reading and writing to local project markdown files. It does not possess network access, shell command execution, or environment variable access capabilities.
Sanitization: There is no explicit validation or escaping of the content found within the issue table before it is displayed or processed.
[DATA_EXFILTRATION]: No network activity was identified. The skill does not use tools like curl or wget and restricts its file operations to local project-specific documentation.
[COMMAND_EXECUTION]: The skill does not execute arbitrary shell commands or attempt privilege escalation. All mentioned workflows refer to standard agent-led file modifications of markdown documentation.

i-have-adhd