The Agent Skills Directory

[SAFE]: The skill exclusively defines linguistic and formatting rules for writing prose. No malicious intent, obfuscation, or sensitive data access was detected.
[COMMAND_EXECUTION]: The skill mentions the use of a local shell script scripts/lint-prose.sh for linting drafts. As this script is a local resource provided by the author to support the skill's primary functionality, it is considered a legitimate developer tool and safe in this context.
[PROMPT_INJECTION]: The skill's purpose is to process and transform user-provided prose, creating a surface for indirect prompt injection. However, the associated risk is negligible due to the limited scope of the skill's actions and the absence of dangerous capabilities. \n- Ingestion points: User-supplied prose to be edited, drafted, or revised (SKILL.md). \n- Boundary markers: No specific delimiters or instructions to ignore embedded commands are present. \n- Capability inventory: Text generation and execution of a local linting script (SKILL.md). \n- Sanitization: No evidence of input validation or sanitization for processed content.

research-writing