The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell scripts (phase1_discovery.sh and phase2_analysis.sh) to perform static analysis of the codebase. These scripts use standard tools like find, grep, wc, and git to collect metrics and identify potential issues.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the analyzed project directory.
Ingestion points: Files are read and searched using cat, grep, and find in the provided scripts.
Boundary markers: None present; the skill does not use specific delimiters to separate untrusted code content from its own instructions during report generation.
Capability inventory: The skill has capabilities to read files, search the directory structure, and write markdown reports to docs/analyze/.
Sanitization: None detected; content found in the codebase is directly included in the generated reports.

proje-analizcisi