tegwick
883409665f
HF-WP-0005. The prompts used to drive work across HelixForge repositories existed only as text pasted between sessions. They are now Canned Prompt Format v0.2 packages with declared inputs, parameters and output contracts. Seven packages: repo-orient (merging "what is this repo about" with "what should we do now" as a `depth` parameter), repo-register, repo-advance, commit-sync, scope-audit, gap-workplan, session-close. "Go on implement, please" is deliberately not packaged. It is conversational continuation with no contract to declare, and packaging it would produce an artifact whose only content is the word "continue". Every one of these prompts assumed the operator's setup — the State Hub is a read model, workplans originate as files and are never registered by hand, a session closes with a progress event. That assumption is what made them personal rather than reusable. helix/custodian-conventions states those rules once as a fragment composed by every package, so they are versioned, improvable in one place, and present for an agent that has never seen this fleet. helix/commit-sync-routine exists for a narrower reason. repo-advance first composed the whole commit-sync package, which itself composes the conventions, so the rendered prompt carried the conventions block twice — CPF inclusion does not deduplicate, and a diamond dependency renders shared content once per path. Factoring the routine out removes the diamond and is a better factoring regardless. Recorded upstream as canned-prompts CANP-WP-0004-T03. helix/session-close composes practice/pqrst-estimate, so the canonical PQRST block is embedded verbatim. hall-of-helix CLOSING.md says the canonical prompt still governs when you have no pqrst-practice checkout, without saying how it reaches you; this is how. evals/pqrst-embedded.yaml fails if the canonical text stops being embedded verbatim, rather than letting the package quietly become a paraphrase. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM Assistant: claude-code Assistant-Model: opus Assistant-Process: 388925@bnt-lap001 Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
1.8 KiB
{{ conventions }}
Task
Orient yourself in this repository and tell me what to do next.
Answer at {{ depth }} depth.
full— work through all four sections below, then recommend.recommendation— do the same reading, but report only the recommendation and the reasoning that supports it. Brevity here is about output, not effort: do not skip the reading.
1. What is this repo for
State its purpose in your own words from INTENT.md, SCOPE.md and the code —
not by quoting the README. If intent and implementation disagree, say so; that
disagreement is usually the most useful thing you will find.
2. What work is pending
List open workplans and their unfinished tasks, from the workplan files rather
than from memory. For each, give its status and what is actually blocking it.
Note any task marked ready whose premises no longer hold.
3. What completing it is worth
For each significant piece of pending work, say what changes once it is done — a capability gained, a risk retired, a manual step removed, a decision unblocked. Be concrete. "Improves quality" is not a benefit; "removes the manual reconciliation step that currently precedes every release" is.
Where the benefit is speculative, say so rather than inflating it.
4. What it would cost
Estimate effort per item as small (under a session), medium (a session or two), or large (multiple sessions, or needs a decision first). Name the main uncertainty driving each estimate. Do not give hour figures — they imply a precision you do not have.
5. Recommendation
Recommend one next action, and say why it beats the alternatives you considered. Prefer work that unblocks other work, retires a risk, or resolves a decision others are waiting on. If the honest answer is that nothing here is worth doing now, say that.