The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/probe_implicit_token_savings.py uses subprocess.run and a helper run_shell to execute system tools such as ls, tree, ripgrep, git, and docker. These commands are used to verify tool availability and output format against temporary local fixtures.\n- [DYNAMIC_EXECUTION]: The regression test suite in scripts/test_probe_implicit_token_savings.py dynamically generates an executable shell script and modifies the process PATH to mock the behavior of external tools (specifically docker) for verification purposes.\n- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by processing repository metadata and contents via exploration tools.\n
Ingestion points: Untrusted data enters the agent context through the output of commands like rg, tree, git log, git diff, and ls.\n
Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from following instructions embedded within these outputs.\n
Capability inventory: The skill includes scripts that execute shell commands and interact with the file system via subprocess.\n
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the repository before it is presented to the agent.

implicit-token-savings