Explains codebase architecture and subsystem flows using parallel explorer and critic agents.
npx skills add https://github.com/adjohn/pstack --skill how-adjohn
Please help me install this Agent Skill. Skill: how Source: https://github.com/adjohn/pstack/tree/main/skills/how Command: npx skills add https://github.com/adjohn/pstack --skill how-adjohn
High-intent search queries and answers about installing and using this skill.
Ask a "how does X work" question and the skill spawns parallel read-only explorer agents that trace entry points, call chains, and data flow. A synthesizer agent then merges their findings into a structured explanation with overview, key concepts, flow, file map, and gotchas.
Ask for architectural issues or improvements rather than just understanding. The skill first produces a full explanation, then spawns one critic per configured model tier that reads the actual code and reports findings with severity, evidence, and impact.
Yes. Narrow questions skip the multi-explorer fan-out entirely. A single agent explores and explains in one pass, which keeps simple questions fast while complex subsystems still get the full parallel exploration treatment.
No. Every explorer, explainer, and critic agent is explicitly instructed to be read-only, forbidding file writes and mutating commands. The agents only use Glob, Grep, and Read to inspect the codebase.
Use it for cross-cutting subsystems spanning multiple files or services where manual tracing is slow. For a single small function you already have open, reading directly is faster; the skill itself leans toward the simple path when in doubt.
sefllessGenerate Markdown walkthroughs from GitHub repositories with code snippets.
radenadriExplain mental models and actionable planning steps before coding in /lmf workflows.
TibsfoxTraces the lineage of decisions and requirements across GSD artifacts.
TokenRollAIInvestigate codebase questions using llmdoc documentation and report findings.
