The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions suggest that users provide sensitive Ethereum private keys as command-line arguments (e.g., git-escrows new-client --privateKey "0x..."). This practice can expose credentials in shell history, system logs, and process monitors.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute various commands, including git, npm, and the git-escrows CLI, based on parameters gathered from the environment and user input.
[EXTERNAL_DOWNLOADS]: The skill encourages the installation of a third-party package (git-escrows) from the public npm registry.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Git repositories and user-supplied URLs which are then used to construct shell commands.
Ingestion points: Git repository URLs and commit hashes are ingested via the Bash and Read tools as described in SKILL.md.
Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the ingested data.
Capability inventory: The skill utilizes Bash for command execution and Read for file access.
Sanitization: No sanitization or validation logic is applied to the repository URLs or commit hashes before they are interpolated into shell commands.

make-git-escrow