The Agent Skills Directory

[SAFE]: The skill uses local file operations (Read, Write, Glob, Grep) solely to create and manage the generated agreement documents within the agent's environment. All operations are local and triggered by explicit user requests for document generation.
[SAFE]: All external references provided in the resources section are links to established legal standard organizations (such as CommonPaper, Bonterms, and ICC) or public open-source template repositories (Open-Agreements) used for informational benchmarking. No remote scripts or executable code are downloaded or invoked from these locations.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied information to populate agreement templates, creating a potential surface for indirect injection if the resulting files are processed by other agents.
Ingestion points: User inputs collected via the information-gathering wizard described in SKILL.md (e.g., party names, addresses, payment terms).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the user-provided data interpolated into the Markdown output.
Capability inventory: The skill possesses Write and Read tools, which are necessary for its primary function of creating agreement files.
Sanitization: There are no explicit instructions to sanitize or validate the user-provided strings before they are written to the output file.
[SAFE]: The skill includes a prominent legal disclaimer that clarifies the tool is for informational purposes and does not establish an attorney-client relationship, which is a best practice for legal-adjacent AI utilities.

agreement-generator