mirror of
https://github.com/J3vb/OwnCord.git
synced 2026-09-03 03:50:00 +03:00
* chore: track the graphify knowledge graph and the bug-hunt ledger Both were local-only, so a clone — including a cloud session, which sees only tracked files — started with no graph and no findings history. graphify-out/: the top-level built graph is now tracked so a fresh clone can query it without a rebuild. Subdirectories stay ignored: cache/ is a per-machine AST cache, and graphify parks the previous graph in a dated YYYY-MM-DD/ backup on every rebuild (18 MB of stale duplicate, a local rollback aid rather than shared state). .gitattributes marks the tree -text: the repo-wide `* text=auto eol=lf` rule would otherwise rewrite line endings inside .graphify_labels.json.sig, which signs the labels byte-for-byte, and invalidate the signature on checkout. graph.json/graph.html also get -diff, and the tree is linguist-generated so it stays out of language stats and collapses in review. .superpowers/: findings-ledger.json, its FINDINGS.md render and render-ledger.mjs are tracked so contributors can add findings by PR. Hunt transcripts, .bak snapshots and debris patches remain per-session scratch. Tradeoff accepted deliberately: the post-commit rebuild hook rewrites graph.json, so each refresh writes a fresh ~18 MB blob into history. Refresh it in its own commit rather than folding it into an unrelated diff. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(graphify): refresh the graph over the newly-tracked files The first commit added findings-ledger.json, FINDINGS.md and render-ledger.mjs to the tracked tree, so the post-commit rebuild picked them up and rewrote the graph. Also ignores .pending_changes, the transient rebuild-state file. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(graphify): share the PreToolUse graph-first nudge hooks The two hook-guard hooks lived in the gitignored settings.local.json with an absolute C:/Users path, so no other clone got them. Portable form: bare `graphify` off PATH, and `|| exit 0` so a contributor without graphify installed is never blocked. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * Update graph output files and manifest - Updated binary files: graph.html and graph.json with new content. - Added new entry for findings-ledger.json in manifest.json with updated metadata. --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
120 lines
2.9 KiB
Plaintext
120 lines
2.9 KiB
Plaintext
# Docker secrets
|
|
.env
|
|
Server/.env
|
|
|
|
# Claude Code — local-only by default. The exceptions are committed on purpose:
|
|
# a cloud session clones this repo and sees ONLY tracked files, so the CLAUDE.md
|
|
# files, skills and workflows have to be here or it starts with no instructions.
|
|
# Machine-local state (settings.local.json, locks) stays ignored.
|
|
.claude/*
|
|
!.claude/skills/
|
|
!.claude/workflows/
|
|
!.claude/settings.json
|
|
CLAUDE.local.md
|
|
.mcp.json
|
|
|
|
# GitHub Copilot instructions (internal)
|
|
.github/copilot-instructions.md
|
|
.github/instructions/
|
|
|
|
# Agent worktrees (Mission Control fleet isolation)
|
|
.worktrees/
|
|
|
|
# AI-specific / internal planning docs
|
|
docs/brain/
|
|
docs/CODEMAPS/
|
|
docs/designs/
|
|
docs/research/
|
|
docs/superpowers/
|
|
/skills/
|
|
|
|
# Mutation-testing output (npm run test:mutate). Local-only by design: a
|
|
# surviving-mutant report maps exactly which behaviour nothing tests.
|
|
Client/tauri-client/.stryker-tmp/
|
|
Client/tauri-client/reports/
|
|
|
|
# Server runtime artifacts
|
|
Server/chatserver.exe
|
|
Server/chatserver.exe~
|
|
Server/owncord-server.exe
|
|
Server/server.exe
|
|
Server/config.yaml
|
|
Server/data/
|
|
|
|
# Test coverage artifacts
|
|
*.out
|
|
Server/cov.out
|
|
Server/cover.out
|
|
Server/coverage.out
|
|
Server/ws_cover.out
|
|
Server/ws_cov.out
|
|
|
|
# Client build artifacts
|
|
Client/publish/
|
|
Client/publish-single/
|
|
Client/publish-release/
|
|
|
|
# HTML mockups (large design reference files)
|
|
Client/login-mockup.html
|
|
Client/ui-mockup.html
|
|
|
|
# Tauri typegen (auto-generated IPC bindings)
|
|
Client/tauri-client/src/generated/
|
|
.typecache
|
|
|
|
# Node modules
|
|
node_modules/
|
|
|
|
# AI tooling
|
|
.gstack/
|
|
.claude-flow/
|
|
.rust-review-results/
|
|
|
|
# Bug-hunt ledger: shared so contributors can add findings. Only the ledger,
|
|
# its render and its validator are tracked; hunt transcripts, .bak snapshots
|
|
# and debris patches are per-session scratch and stay local.
|
|
.superpowers/*
|
|
!.superpowers/findings-ledger.json
|
|
!.superpowers/FINDINGS.md
|
|
!.superpowers/render-ledger.mjs
|
|
|
|
.claude/worktrees/
|
|
|
|
# Internal dev tools (e.g. tools/livekit-server.exe) are ignored, but the
|
|
# owncord-introspect MCP server is a committed, shared dev tool.
|
|
tools/*
|
|
!tools/mcp-introspect/
|
|
tools/mcp-introspect/node_modules/
|
|
.cache/
|
|
|
|
# Internal dev files (root-level scratch only)
|
|
/SKILL.md
|
|
TODOS.md
|
|
DESIGN.md
|
|
Client/CLIENT-REVIEW.md
|
|
|
|
# Agent tooling state
|
|
.serena/
|
|
|
|
# Client env (holds API keys - never commit)
|
|
Client/tauri-client/.env
|
|
|
|
# Rust review output
|
|
.rust-review-results/
|
|
|
|
# Claude Code worktrees (local scratch, never commit)
|
|
.claude/worktrees/
|
|
|
|
# local server run logs
|
|
server.log
|
|
|
|
# Knowledge graph (graphify). The top-level built graph is tracked so a fresh
|
|
# clone can query it without a rebuild. Every subdirectory stays local: cache/
|
|
# is a per-machine AST build cache, and graphify parks the PREVIOUS graph in a
|
|
# dated YYYY-MM-DD/ backup on each rebuild — 18 MB of stale duplicate that is a
|
|
# local rollback aid, not shared state.
|
|
graphify-out/*/
|
|
|
|
# Transient graphify rebuild-state file.
|
|
graphify-out/.pending_changes
|