System Prompt (SOP)
The SOP is the highest-leverage AI artifact in this setup. It decides how every harness interprets a request before a skill, subagent, model, or workflow gets involved.
Read these pages as human documentation about the system's contract: where the prompt lives, what it requires, and how the surrounding skills and workflows layer on top.
The moving parts
| Slice | Owns |
|---|---|
| Source of truth | single file, symlink fan-out, and update workflow |
| Truth and verification | compatibility, external truth, runtime truth, completion, compact-output recovery, self-report skepticism |
| Execution workflow | reverse interview, persistent specs, verification loops, state-machine harnesses |
| Side-effect gates | git push safety, ownership, publication, bot/human split |
| Code-quality and dotfiles policy | style matching, semantic dedupe, docs hygiene, shell/helper rules |
Using it
Start with Source of truth when changing the SOP itself. Use the other slices to understand the rule family a harness, skill, or review workflow is enforcing.
The mental model is layered:
- SOP sets global invariants.
- Skills add intent-specific procedures.
- Subagents isolate heavy or parallel work.
- Workflow pages explain the common composed flows.
Source map
| Surface | Source | Target |
|---|---|---|
| Single SOP source | home/readonly_AGENTS.md | ~/AGENTS.md |
| Claude | home/symlink_CLAUDE.md | ~/CLAUDE.md |
| Gemini | home/dot_gemini/symlink_GEMINI.md | ~/.gemini/GEMINI.md |
| Cursor | home/dot_cursor/symlink_AGENTS.md | ~/.cursor/AGENTS.md |
| Codex | home/dot_codex/symlink_AGENTS.md | ~/.codex/AGENTS.md |
| OpenCode | home/dot_config/opencode/symlink_AGENTS.md | ~/.config/opencode/AGENTS.md |
| Copilot | home/private_dot_copilot/symlink_copilot-instructions.md | ~/.copilot/copilot-instructions.md |