docs(positioning): agent-native narrative — canonical positioning page, README restructure, contributor setup skill#142
Merged
Conversation
…it-body wrap rule Completes the remaining #134 work. - .github/workflows/pr-title.yml lints the PR title (which becomes the squash-merge commit subject) as an English Conventional Commit, reusing scripts/hooks/commit_msg_check.py so the convention keeps one source. This is the server-side floor for external contributors, whom the local commit-msg git hook cannot bind because it only fires after `pre-commit install`. - contexts/README.md now states the contexts/ vs docs/ boundary explicitly: contexts/ is internal navigation and design memory for agents and maintainers; docs/ is the user-facing catalog and guides. This was the last open acceptance criterion in #134. - commit.md (both skill copies) adds the commit-body convention: follow the same no-hard-wrap prose rule as GitHub bodies, so a squash-merge keeps a clean description. This commit message follows it — one physical line per list item. - contexts/dev/harness-engineering.md records the PR-title lint in the enforcement table and notes that making it a required status check is a one-time repo-settings toggle. verify.sh passes (426 passed, 85.65% coverage). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…/ link The commit-body rule referenced github-writing.md via `../../../../contexts/dev/github-writing.md`, matching the repo's tested clickable-link convention but reading badly. commit.md is not test-bound to that link, so use a readable repo-root-relative backtick reference instead. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…with data-mono) Skill files referenced contexts pages with deep, fragile clickable links like `../../../../contexts/dev/github-writing.md`. The sister repo llmquant-data-mono instead uses repo-root-relative backtick references (`contexts/...`) from nested files, because a clean clickable repo-root link is only possible from a root-level file — GitHub resolves a bare relative link against the current file's directory, not the repo root, so a nested file needs the ugly `../` chain to stay clickable. - SKILL.md and pull-request.md (both skill copies) now reference contexts pages as repo-root backtick paths (`contexts/README.md`, `contexts/dev/labels.md`, `contexts/dev/github-writing.md`) instead of `../../../` links. Root-level files (AGENTS.md) and sibling contexts pages keep their clean clickable links. - tests/test_contexts.py: the three route tests now check that each source references the canonical guide in the form appropriate to its location — a backtick repo-root reference for nested skill files, a clickable link for root or sibling pages — and that the guide exists at the repo root, instead of asserting a specific deep relative link path. verify.sh passes (426 passed, 85.65% coverage). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…estructure, contributor setup - contexts/design/positioning.md: canonical positioning source (context engineering / harness engineering as dimensions of one repo; V1/V2 kept only as a poster-mapping note), registered in the design index and CONTEXT_MAP - README.md: workbench-first restructure — original slogan restored, Knowledge Engineering + The Vision sections with figures, agent-path quick start, PaperSemanticCfg examples, design-phase evaluation NOTE with the Anthropic evals reference, LLMQuant Data screenshot, direction-based roadmap, agent-native contributing prompts - AGENTS.md: positioning paragraph aligned to the dimension names - quantmind-dev skill: new references/setup.md (contributor setup + issue filing) with SKILL.md routing, mirrored byte-identically to .agents - github-writing: no-hard-wrap rule extended to README.md Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
keli-wen
added
type: docs
Resolve README.md quick-start conflict: keep the dual-cfg PaperFlow examples (structure tree + semantic, with master's real result fields) and adopt master's batch_run and magic examples. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
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
Aligns the repository's narrative with the converged July 2026 positioning: QuantMind is an agent-native workbench for financial knowledge extraction, structured by two engineering dimensions — context engineering (any source → typed knowledge) and harness engineering (any agent → domain specialist). This PR lands the canonical positioning page, restructures the README around that story, and makes the README's agent-native contributing prompts real.
Changes
contexts/design/positioning.md(new) — the single canonical positioning source: positioning and hero, the two dimensions, the bet, evaluation (design phase, no results claimed), the LLMQuant Data relationship, and the roadmap. The V1/V2 labels are documented as a poster-only contrast device; serious prose uses the dimension names. Registered in the design index andCONTEXT_MAP.md. Supersedes the April positioning spec.README.md— workbench-first restructure: original slogan restored; Knowledge Engineering and The Vision sections with the two architecture figures; agent-path-first Quick Start withPaperStructureCfg/PaperSemanticCfgexamples on onePaperFlow; a design-phase evaluation[!NOTE]referencing Anthropic's "Demystifying evals for AI agents"; LLMQuant Data production section with screenshot; direction-based roadmap; agent-native contributing prompts. Prose is unwrapped (no fixed-width hard wrapping).AGENTS.md— positioning paragraph aligned to the dimension names and linked to the canonical page.quantmind-devskill — newreferences/setup.md(contributor setup and issue filing) withSKILL.mdrouting so the README prompts (set me up as a contributor,file an issue,contribute a change) resolve; mirrored byte-identically under.agents/.contexts/dev/github-writing.md— the no-hard-wrap prose rule now also coversREADME.md.This branch also carries two pending follow-up commits from
feat/harness-close-134-pr-title-and-boundary(repo-root-relative cross-references in skills and contexts) that had no open PR of their own.Verification
bash scripts/verify.shgreen (ran in the pre-push hook);tests/test_contexts.pypasses with the new page registered. README example symbols (PaperFlow,PaperStructureCfg,PaperSemanticCfg,collect_news,batch_run) verified againstquantmind/flowsandquantmind/configsexports on master after #140.🤖 Generated with Claude Code