The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses system shell commands (xdg-open, open, and start) to automatically launch the generated HTML report in the user's default browser or viewer.
[PROMPT_INJECTION]: The skill ingests untrusted data from the local workspace (e.g., CLAUDE.md, memory/ folders, and srd/ project files) to ground its analysis. This content is passed to sub-agents without explicit sanitization, which could lead to indirect prompt injection if the files contain malicious instructions.
Ingestion points: Local workspace files including CLAUDE.md, memory/ directories, and Linear-related srd/ folders.
Boundary markers: None identified in the prompt template used for sub-agents.
Capability inventory: The skill can spawn new sub-agents using the Agent tool and execute local shell commands to open files.
Sanitization: The skill does not perform escaping or filtering of the workspace content before interpolating it into prompts.
[EXTERNAL_DOWNLOADS]: The skill's documentation provides links to third-party GitHub repositories (expectedparrot/premortem and iepathos/premortem) referenced as prior art.

premortem