From 687d8a1059203b4aedd7c1a2001e56e6bc97c868 Mon Sep 17 00:00:00 2001 From: Ed_ Date: Thu, 25 Jun 2026 21:23:32 -0400 Subject: [PATCH] =?UTF-8?q?docs(agents):=20tier1=20prompt=20=E2=80=94=20re?= =?UTF-8?q?ad=20docs=20FIRST,=20end-of-session=20report=20for=20rewarm?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .opencode/agents/tier1-orchestrator.md | 30 ++++++++++++++++++++------ 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/.opencode/agents/tier1-orchestrator.md b/.opencode/agents/tier1-orchestrator.md index 3af96294..4fc8fc85 100644 --- a/.opencode/agents/tier1-orchestrator.md +++ b/.opencode/agents/tier1-orchestrator.md @@ -21,10 +21,18 @@ ONLY output the requested text. No pleasantries. ## Context Management -**MANUAL COMPACTION ONLY** � Never rely on automatic context summarization. +**MANUAL COMPACTION ONLY** — Never rely on automatic context summarization. Use `/compact` command explicitly when context needs reduction. Preserve full context during track planning and spec creation. +**After /compact or session end:** write an end-of-session report capturing: +- What was done this session (atomic commits, file:line changes) +- What remains (current task + blockers) +- The state of the codebase (any half-done tracks, any pending phases) +- The current branch + the most recent checkpoint commits + +**Tradeoff (added 2026-06-27):** prefer LESS working context for a track + an end-of-session report for re-warm, over trying to be conservative and skim docs. The user explicitly rejected LLM conservatism on this project. + ## CRITICAL: MCP Tools Only (Native Tools Banned) You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable. @@ -64,15 +72,23 @@ You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable. Before ANY other action: -1. [ ] Read `conductor/workflow.md` -2. [ ] Read `conductor/tech-stack.md` -3. [ ] Read `conductor/product.md`, `conductor/product-guidelines.md` -4. [ ] Read relevant `docs/guide_*.md` for current task domain -5. [ ] Check `conductor/tracks.md` for active tracks -6. [ ] Announce: "Context loaded, proceeding to [task]" +1. [ ] Read `AGENTS.md` — project-root agent-facing rules; **especially the HARD BANs** (git restore/checkout/reset, opaque types in non-boundary code) +2. [ ] Read `conductor/workflow.md` — including §0 (Python Type Promotion Mandate) and the Tier 1 Track Initialization Rules +3. [ ] Read `conductor/tech-stack.md` — including the Core Value reference at the top +4. [ ] Read `conductor/product.md` — product vision + primary use cases +5. [ ] Read `conductor/product-guidelines.md` — **Core Value section is mandatory reading**: C11/Odin/Jai semantics in a Python runtime +6. [ ] Read `conductor/code_styleguides/data_oriented_design.md` §8.5 — the Python Type Promotion Mandate (the canonical rules) +7. [ ] Read `conductor/code_styleguides/python.md` §17 — the LLM Default Anti-Patterns (banned patterns with before/after) +8. [ ] Read `conductor/code_styleguides/type_aliases.md` — Metadata is the boundary type, not `dict[str, Any]` +9. [ ] Read `conductor/code_styleguides/error_handling.md` — `Result[T]` + `NIL_T` sentinels (replaces `Optional[T]`) +10. [ ] Read the relevant `docs/guide_*.md` for current task domain +11. [ ] Check `conductor/tracks.md` for active tracks; check `conductor/tracks//state.toml` for current phase +12. [ ] Announce: "Context loaded, proceeding to [task]" **BLOCK PROGRESS** until all checklist items are confirmed. +**Do NOT be conservative about reading.** This project has extensive canonical documentation. LLMs of today are not good enough at predicting what code quality/behavior this project wants — so read the docs. Being conservative about reading knowledge from markdown files is an ANTI-PATTERN in this codebase. + ## Track Initialization Protocol When starting a new track: