The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted external files.
Ingestion points: The skill uses the Read tool to ingest the content of a target file in SKILL.md and phases/01-analysis.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious embedded instructions within the files being analyzed.
Capability inventory: The skill utilizes powerful tools including Read, Write, Edit, Bash, Glob, and Grep across its execution phases.
Sanitization: The instructions do not define any sanitization, escaping, or validation logic for the content read from the external files before it is processed or used in analysis results.
[COMMAND_EXECUTION]: While the skill includes Bash in its allowed-tools and identifies bash command blocks for optimization, the logic provided is restricted to text-based refactoring and does not demonstrate the execution of arbitrary or attacker-controlled shell commands.
[DATA_EXFILTRATION]: The skill performs local file read and write operations on target markdown files. It does not contain any network-enabled tools or patterns suggesting the transmission of data to external domains.

skill-simplify