The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external web sources and local project files to generate GTM plans, creating an attack surface where malicious instructions in those sources could influence agent behavior.
Ingestion points: The skill utilizes WebSearch and WebFetch for external research, and reads local files including README.md, package.json, and project documentation (docs/**/*.md) to gather strategy context.
Boundary markers: The prompts provided to sub-agents (paid-research, unpaid-research, etc.) lack explicit delimiters or instructions to ignore commands that may be embedded within the gathered data.
Capability inventory: The skill possesses capabilities to execute shell commands via Bash, modify local files via Write, and invoke other agent capabilities through the Skill tool (delegating to ops-marketing).
Sanitization: No sanitization, escaping, or schema validation is performed on external data before it is interpolated into prompts or used to define campaign parameters.
[COMMAND_EXECUTION]: The skill uses the Bash tool to perform reconnaissance on the local system and repository to automate the setup process.
Evidence: The setup and Repo auto-scan routines execute commands such as git remote -v, jq, and cat to extract information about the current project and user preferences.

ops-gtm