Migrate .codeflash/ to {teammember}/{org}/{project}/ format (#15)

Add team member dimension to case study paths so multiple contributors
can track optimization data independently. Derives member from
git config user.name in session-start hooks.

- Move all case studies under .codeflash/krrt7/
- Rename pypa/pip → python/pip (org grouping)
- Update session-start hooks, docs, scripts, and references

.claude/hooks/post-compact.sh

@@ -47,7 +47,7 @@ fi
 STATE="${STATE}\nProject conventions to preserve:\n"
 STATE="${STATE}- Monorepo: packages/ (UV workspace), plugin/ (self-contained, multi-language: plugin/languages/python/, plugin/languages/javascript/)\n"
 STATE="${STATE}- Build: make build-plugin, prek run --all-files (lint), uv run pytest packages/ -v (test)\n"
-STATE="${STATE}- Optimization projects in .codeflash/{org}/{project}/ with status.md, bench/, data/results.tsv\n"
+STATE="${STATE}- Optimization projects in .codeflash/{teammember}/{org}/{project}/ with status.md, bench/, data/results.tsv\n"
 STATE="${STATE}- Target repos in ~/Desktop/work/{org}_org/{project}\n"
 STATE="${STATE}- VM benchmarks via ssh -A, record to data/results.tsv, update status.md\n"
 STATE="${STATE}- Atomic commits, one purpose per commit, verify before committing\n"

.claude/hooks/session-start.sh

@@ -1,22 +1,26 @@
 #!/usr/bin/env bash
-# SessionStart hook: Scaffold .codeflash/{org}/{project}/ if it doesn't exist.
-# Infers org/project from git remote origin. File generation is delegated to
-# scripts/scaffold.sh — the single source of truth for project scaffolding.
+# SessionStart hook: Scaffold .codeflash/{teammember}/{org}/{project}/ if it doesn't exist.
+# Infers org/project from git remote origin. Team member is derived from
+# git user name. File generation is delegated to scripts/scaffold.sh.
 
 cd "$CLAUDE_PROJECT_DIR" 2>/dev/null || exit 0
 
 CF_DIR="$CLAUDE_PROJECT_DIR/.codeflash"
 SCAFFOLD="$CLAUDE_PROJECT_DIR/scripts/scaffold.sh"
 
+# Derive team member from git user (lowercase, no spaces)
+MEMBER=$(git config user.name 2>/dev/null | tr '[:upper:]' '[:lower:]' | tr ' ' '-')
+[ -z "$MEMBER" ] && MEMBER="unknown"
+
 # Parse git remote origin
 REMOTE=$(git remote get-url origin 2>/dev/null)
 if [ -z "$REMOTE" ]; then
     if [ -d "$CF_DIR" ]; then
         exit 0
     fi
-    cat <<'EOF'
+    cat <<EOF
 {
-  "systemMessage": "No .codeflash/ directory found and no git remote origin to infer org/project. Ask the user for the organization and project name, then run: bash scripts/scaffold.sh <org> <project> .codeflash/<org>/<project>"
+  "systemMessage": "No .codeflash/ directory found and no git remote origin to infer org/project. Ask the user for the organization and project name, then run: bash scripts/scaffold.sh <org> <project> .codeflash/$MEMBER/<org>/<project>"
 }
 EOF
     exit 0
@@ -51,15 +55,15 @@ if [ -z "$ORG" ] || [ -z "$PROJECT" ]; then
     if [ -d "$CF_DIR" ]; then
         exit 0
     fi
-    cat <<'EOF'
+    cat <<EOF
 {
-  "systemMessage": "No .codeflash/ directory found. Could not parse org/project from git remote. Ask the user for the organization and project name, then run: bash scripts/scaffold.sh <org> <project> .codeflash/<org>/<project>"
+  "systemMessage": "No .codeflash/ directory found. Could not parse org/project from git remote. Ask the user for the organization and project name, then run: bash scripts/scaffold.sh <org> <project> .codeflash/$MEMBER/<org>/<project>"
 }
 EOF
     exit 0
 fi
 
-PROJECT_DIR="$CF_DIR/$ORG/$PROJECT"
+PROJECT_DIR="$CF_DIR/$MEMBER/$ORG/$PROJECT"
 
 # Skip bootstrap when working on the agent repo itself
 if [ "$ORG" = "codeflash-ai" ] && [ "$PROJECT" = "codeflash-agent" ]; then

.claude/rules/optimization-projects.md

@@ -2,7 +2,7 @@
 
 ## Location
 
-Active optimization data lives in `.codeflash/{org}/{project}/` on main. Summaries are built into `case-studies/{org}/{project}/`.
+Active optimization data lives in `.codeflash/{teammember}/{org}/{project}/` on main. Summaries are built into `case-studies/{org}/{project}/`.
 
 ## Status tracking
 

.codeflash/krrt7/typeagent-py/HANDOFF.md

@@ -0,0 +1,27 @@
+# Optimization PR Stack Handoff
+
+## PR Stack (order of merge)
+
+| PR | Title | Status |
+|----|-------|--------|
+| #231 | perf: Use `executemany` for SQLite batch inserts | Merged |
+| #229 | perf: Batch term collection in indexing pipeline | Merged |
+| #230 | perf: Batch SQLite INSERTs for indexing pipeline | Awaiting re-review |
+| #232 | perf: Batch metadata query | Blocked on #230 |
+
+## What was done in #230
+
+- Addressed all 10 review comments from @bmerkle
+- Deleted `text_range_from_location` (was identical to `text_range_from_message_chunk` from messageutils)
+- Deleted `add_entity_to_index`, `add_action_to_index`, `add_topic_to_index` (duplicates of `add_entity`, `add_action`, `add_topic`)
+- Updated all callers (tests) to use the unified functions directly
+- Fixed pre-existing bug: `inverse_actions` were silently skipped in both the batch path (`add_metadata_to_index_from_list`) and the async iterator path (`add_metadata_to_index`)
+- Moved inline imports to top-level in `sqlite/propindex.py` per AGENTS.md guidelines
+- Summary comment posted on #230, heads-up comment posted on #232
+
+## What's next
+
+1. Wait for @bmerkle / @gvanrossum to re-review #230
+2. Once #230 merges, rebase #232 (`perf/batch-metadata-query`) on `main` and resolve conflicts
+   - #232 will have conflicts in `semrefindex.py` due to our function renames and deduplication
+3. CI checks on #230 were still queued at last check — monitor for results

CLAUDE.md

@@ -4,15 +4,16 @@ Monorepo for the Codeflash optimization platform: Python packages, Claude Code p
 
 ## Case Studies
 
-Active case study data lives in `.codeflash/{org}/{project}/` (status, bench scripts, raw data, VM infra). Summaries are built out of `.codeflash/` into `case-studies/{org}/{project}/`.
+Active case study data lives in `.codeflash/{teammember}/{org}/{project}/` (status, bench scripts, raw data, VM infra). Summaries are built out of `.codeflash/` into `case-studies/{org}/{project}/`.
 
-Active case studies in `.codeflash/`:
+Active case studies in `.codeflash/krrt7/`:
 - `microsoft/typeagent`
 - `unstructured/core-product`
 - `netflix/metaflow`
 - `coveragepy/coveragepy`
 - `textualize/rich`
-- `pypa/pip`
+- `python/pip`
+- `odoo`
 
 ### Directory conventions
 
@@ -28,8 +29,8 @@ Target repos live in `~/Desktop/work/{org}_org/{project}`:
 2. **Run tests locally** to verify nothing breaks
 3. **Commit and push** to the fork
 4. **Benchmark on the VM** via `ssh -A azureuser@<ip> "cd ~/<project> && git fetch origin && ..."`
-5. **Record results** in `.codeflash/{org}/{project}/data/results.tsv`
-6. **Update status.md** in `.codeflash/{org}/{project}/`
+5. **Record results** in `.codeflash/{teammember}/{org}/{project}/data/results.tsv`
+6. **Update status.md** in `.codeflash/{teammember}/{org}/{project}/`
 7. **Open a PR** on the fork with VM benchmark numbers
 
 ### VM access

README.md

@@ -176,11 +176,11 @@ plugin/                        # Claude Code plugin (self-contained, multi-langu
   languages/python/            # Python domain agents, skills, references
   languages/javascript/        # JavaScript domain agents, skills, references
 
-.codeflash/                  # active optimization data (org-grouped)
-  textualize/rich/             # 2x Rich import speedup
-  pypa/pip/                    # 7x pip --version, 1.81x resolver
-  microsoft/typeagent/         # Structured RAG optimization
-  <org>/<project>/             # new optimization targets
+.codeflash/                  # active optimization data (teammember/org/project)
+  krrt7/textualize/rich/       # 2x Rich import speedup
+  krrt7/python/pip/            # 7x pip --version, 1.81x resolver
+  krrt7/microsoft/typeagent/   # Structured RAG optimization
+  <member>/<org>/<project>/    # new optimization targets
 
 case-studies/                  # summaries built from .codeflash/
 scripts/                       # scaffold scripts
@@ -216,7 +216,7 @@ make bootstrap ORG=unstructured PROJECTS="unstructured unstructured-inference co
 This creates:
 
 ```
-.codeflash/<org>/<project>/
+.codeflash/<member>/<org>/<project>/
 ├── README.md              # results, what changed, methodology (from template)
 ├── bench/                 # add your benchmark scripts here
 ├── data/                  # save raw benchmark data here
@@ -243,7 +243,7 @@ The cloud-init template includes examples for Python, Rust, Go, Node.js, and Jav
 Each project gets a `vm-manage.sh` for the benchmark VM:
 
 ```bash
-cd .codeflash/<org>/<project>
+cd .codeflash/<member>/<org>/<project>
 bash infra/vm-manage.sh create    # provision VM with cloud-init
 bash infra/vm-manage.sh bench main  # run benchmarks on a branch
 bash infra/vm-manage.sh ssh       # SSH into VM
@@ -254,5 +254,5 @@ bash infra/vm-manage.sh destroy   # delete everything
 ### Examples
 
 Use the existing projects as templates:
-- [Rich](.codeflash/textualize/rich/) — focused scope, 2 PRs, import + runtime micro-opts
-- [pip](.codeflash/pypa/pip/) — large scope, 122 commits across 8 categories
+- [Rich](.codeflash/krrt7/textualize/rich/) — focused scope, 2 PRs, import + runtime micro-opts
+- [pip](.codeflash/krrt7/python/pip/) — large scope, 122 commits across 8 categories

plugin/hooks/session-start.sh

@@ -3,7 +3,7 @@
 #
 # Two modes:
 #   Dogfood: Plugin loaded from codeflash-agent/dist/ (--plugin-dir).
-#            Redirects .codeflash/ → agent repo's .codeflash/{org}/{project}/ via symlink.
+#            Redirects .codeflash/ → agent repo's .codeflash/{teammember}/{org}/{project}/ via symlink.
 #   Normal:  Plugin installed normally.
 #            Creates .codeflash/ in the project directory.
 
@@ -27,6 +27,10 @@ PROJECT=$(echo "$PATH_PART" | cut -d'/' -f2 | tr '[:upper:]' '[:lower:]')
 
 [ -z "$ORG" ] || [ -z "$PROJECT" ] && exit 0
 
+# Derive team member from git user (lowercase, no spaces)
+MEMBER=$(git config user.name 2>/dev/null | tr '[:upper:]' '[:lower:]' | tr ' ' '-')
+[ -z "$MEMBER" ] && MEMBER="unknown"
+
 # Detect dogfood mode: CLAUDE_PLUGIN_ROOT's parent has .codeflash/ (the agent repo)
 AGENT_REPO=""
 if [ -n "$CLAUDE_PLUGIN_ROOT" ]; then
@@ -38,7 +42,7 @@ fi
 
 if [ -n "$AGENT_REPO" ]; then
     # --- Dogfood mode ---
-    TARGET="$AGENT_REPO/.codeflash/$ORG/$PROJECT"
+    TARGET="$AGENT_REPO/.codeflash/$MEMBER/$ORG/$PROJECT"
     mkdir -p "$TARGET"
 
     if [ ! -e "$CLAUDE_PROJECT_DIR/.codeflash" ]; then

plugin/references/shared/failure-modes.md

@@ -192,7 +192,7 @@ Optimizer resumes: Reads compacted summary, can't find branch name
 ```
 Teammate: Check available context in order:
 
-Option A: Read .codeflash/{org}/{project}/status.md
+Option A: Read .codeflash/{teammember}/{org}/{project}/status.md
           Should say "Current branch: perf/batch-size"
           Should say "Latest: 40% throughput gain, ready for benchmark"
           
@@ -217,7 +217,7 @@ git branch -v
 git log -1 --format=fuller
 
 # Confirm results exist
-ls -la .codeflash/{org}/{project}/data/
+ls -la .codeflash/{teammember}/{org}/{project}/data/
 
 # Read what lead said in last interaction
 # (Look for approval, next steps in task description)
@@ -245,7 +245,7 @@ Now that context restored, proceed normally
 - **Configure PreCompact hook** to snapshot before compaction
 - **Configure SessionStart hook** to inject status after compaction
 - **Keep MEMORY.md updated** with current branch and findings as you work
-- **Update .codeflash/{org}/{project}/status.md** regularly, not just at handoff
+- **Update .codeflash/{teammember}/{org}/{project}/status.md** regularly, not just at handoff
 
 ---
 
@@ -350,8 +350,8 @@ Optimizer: Compacted summary says "bottleneck found" but no numbers
 **Step 1: Check what was actually lost** (2 min)
 ```bash
 # Do the files still exist?
-ls -la .codeflash/{org}/{project}/data/profile.json
-cat .codeflash/{org}/{project}/data/profile.json | head
+ls -la .codeflash/{teammember}/{org}/{project}/data/profile.json
+cat .codeflash/{teammember}/{org}/{project}/data/profile.json | head
 
 # Is it in teammate MEMORY.md?
 grep -A 5 "O(n²)" MEMORY.md

plugin/references/shared/team-structure.md

@@ -180,7 +180,7 @@ Result: Wrong approach, wastes benchmarking time/VM cost, leads to dead-end bran
 | 4 teammates | 60k | 60-120k | 120-180k | Multi-project or multi-layer work |
 | 5 teammates | 75k | 80-150k | 155-225k | Large exploration, competing hypotheses |
 
-**Cost tracking rule**: If team approach > lead-only, switch back to single session. Use `.codeflash/{org}/{project}/metrics.tsv` to track.
+**Cost tracking rule**: If team approach > lead-only, switch back to single session. Use `.codeflash/{teammember}/{org}/{project}/metrics.tsv` to track.
 
 ## Decision Tree
 

scripts/scaffold.sh

@@ -7,7 +7,7 @@
 #
 # Called by:
 #   - make bootstrap ORG=roboflow PROJECTS="supervision"
-#   - .claude/hooks/session-start.sh (auto-scaffolds .codeflash/{org}/{project}/)
+#   - .claude/hooks/session-start.sh (auto-scaffolds .codeflash/{teammember}/{org}/{project}/)
 
 set -euo pipefail