The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses various shell utilities including grep, find, cat, and python3 to perform static analysis of the codebase. These commands are scoped to identification and scoring of project patterns (e.g., line counts, script existence) and do not include dangerous operations or network access.\n- [DATA_EXPOSURE]: Phase 4 of the analysis contains patterns intended to identify hardcoded secrets and security vulnerabilities within the project's source code. This is an auditing feature for the user; the results are handled locally and no data exfiltration logic was detected.\n- [INDIRECT_PROMPT_INJECTION]: Because the skill processes local source files that may be attacker-controlled, there is a theoretical surface for indirect prompt injection. However, the risk is mitigated as the analysis logic focuses on regex pattern matching and statistical counting rather than interpreting file contents as high-level instructions.

upgrade-analysis