The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a bundled shell script (prepare.sh) to initialize the output directory and validate parameters. It also uses the system's open command to launch the generated HTML report in a browser. The slug parameter is strictly validated against a kebab-case regular expression, mitigating shell injection risks.
[EXTERNAL_DOWNLOADS]: The skill instructions and error messages reference an external dependency, html-playground, providing instructions to install it from the author's GitHub repository (PaulRBerg/agent-skills). This is documented as a necessary component for the skill's template system.
[PROMPT_INJECTION]: The skill operates by ingesting data from the current task transcript to populate HTML reports. As it processes potentially untrusted external data, it presents a surface for indirect prompt injection, although instructions mandate that the agent verify all claims against verified tool evidence.

html-debrief