refactor(ce-brainstorm): dedupe platform refs and extract rigor probes#821
Open
tmchow wants to merge 1 commit into
Open
refactor(ce-brainstorm): dedupe platform refs and extract rigor probes#821tmchow wants to merge 1 commit into
tmchow wants to merge 1 commit into
Conversation
Consolidate the platform blocking-tool enumeration so it lives in one canonical spot (Interaction Rule 4 in SKILL.md) and have references point back via plain prose. Reframe headless-mode framing so Claude Code's `disable-model-invocation` flag is named as an example, not the load-bearing condition. Move Phase 1.2 Standard/Deep/Deep-product gap definitions and Phase 1.3 example probe wordings to a new references/rigor-probes.md so tier-conditional content loads on demand.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
ce-brainstorm now has a single canonical source for per-platform blocking-tool names, portable headless-mode framing, and tier-conditional content that loads on demand instead of on every invocation. SKILL.md shrinks from about 3,490 to 2,990 words while preserving every load-bearing rule.
What changed
The platform blocking-tool enumeration (
AskUserQuestionin Claude Code,request_user_inputin Codex,ask_userin Gemini,ask_userin Pi) was duplicated across SKILL.md and three reference files. It now lives only in Interaction Rule 4 of SKILL.md; the references point back via plain prose ("per Interaction Rule 4 in SKILL.md"). Reference files stop leaking Claude Code awareness, and drift risk drops from four locations to one.Headless-mode framing in SKILL.md and
references/synthesis-summary.mdnow leads with "no synchronous user / non-interactive context" as the load-bearing condition. Claude Code'sdisable-model-invocationfrontmatter flag is named as one example, not the trigger, so the rule still fires correctly on automated flows from Codex, Gemini, Pi, or Kiro that have no equivalent flag.Phase 1.2 Standard/Deep/Deep-product gap definitions and Phase 1.3 example probe wordings moved to a new
references/rigor-probes.md. SKILL.md keeps the framing, the Lightweight set, and the load-bearing behavioral contract (probes are prose, fire before Phase 2, attachment last). Tier-conditional content now loads on demand rather than on every brainstorm.