The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch mandatory reference files and contracts from the author's GitHub repository (github.com/levnikolaevich/claude-code-skills) if they are not found in the local file system.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform file searches and auditing operations.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from external project documents that could contain malicious instructions designed to influence the agent's behavior during the audit.
Ingestion points: Audited project files including AGENTS.md, CLAUDE.md, and various documents within the docs/ directory.
Boundary markers: Absent; the skill lacks explicit delimiters or instructions to ignore embedded commands within the audited text.
Capability inventory: Includes the ability to execute shell commands via Bash and perform file system operations using Read, Grep, and Glob tools.
Sanitization: No sanitization or structural validation is performed on the ingested document content before it is analyzed.

ln-612-semantic-content-auditor