@raquezha/norpiv 0.0.3 → 0.0.5

package/README.md

@@ -24,6 +24,19 @@ Auxiliary hygiene:
 
 ---
 
+## notrace relationship
+
+notrace is an optional retrospective layer around RPIV.
+
+Rules:
+
+- **RPIV must work without notrace**.
+- **notrace may attach to RPIV when task state exists**.
+- `WORK.md` remains the RPIV source of truth whether notrace is installed or not.
+- `.notrace/` owns notrace artifacts.
+- `.workflow/` owns RPIV task state.
+- notrace artifacts and reviews may be referenced from `WORK.md [LOG]`, but RPIV phases must not depend on those files being present.
+
 ## 🛡️ Critical Guardrails
 
 - **Measure Twice, Cut Once**: Never implement code during scoping or planning. The agent will wait for an explicit `EXECUTE` statement before modifying files.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@raquezha/norpiv",
-  "version": "0.0.3",
+  "version": "0.0.5",
   "description": "Bundled RPIV workflow skills for Pi, Claude, and Codex-style coding agents",
   "type": "commonjs",
   "bin": {

package/scripts/get-pi-model.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+# Deterministically grabs the active Pi model from the current session logs
+
+# Format PWD to match Pi's session directory naming convention
+SAFE_PWD=$(pwd | sed 's/\//-/g' | sed 's/^-/--/')
+TARGET_DIR="$HOME/.pi/agent/sessions/${SAFE_PWD}--"
+
+if [ -d "$TARGET_DIR" ]; then
+    LATEST_FILE=$(ls -t "$TARGET_DIR" | head -n 1 2>/dev/null)
+    if [ -n "$LATEST_FILE" ]; then
+        PROVIDER=$(cat "$TARGET_DIR/$LATEST_FILE" | grep -o '"provider":"[^"]*"' | tail -n 1 | cut -d'"' -f4)
+        MODEL=$(cat "$TARGET_DIR/$LATEST_FILE" | grep -o '"model":"[^"]*"' | tail -n 1 | cut -d'"' -f4)
+        
+        if [ -n "$MODEL" ]; then
+            # If provider exists, prefix it, otherwise just return model
+            if [ -n "$PROVIDER" ]; then
+                echo "$PROVIDER:$MODEL"
+            else
+                echo "UnknownProvider:$MODEL"
+            fi
+            exit 0
+        fi
+    fi
+fi
+
+# Fallback if no active session log can be parsed
+echo "UnknownAgent:UnknownModel"

package/scripts/reposcry-bootstrap.sh

@@ -42,6 +42,30 @@ reposcry_indexed_files() {
   reposcry stats 2>/dev/null | awk '/Files indexed:/ {print $3; exit}'
 }
 
+# Quick health check for Pulse info
+check_pulse() {
+  if ! command -v reposcry >/dev/null 2>&1; then
+    echo "PULSE: Missing (Not installed)"
+    return
+  fi
+  if [[ ! -d .reposcry ]]; then
+    echo "PULSE: Cold (No cache)"
+    return
+  fi
+  local files
+  files=$(reposcry_indexed_files || echo "0")
+  if [[ "$files" == "0" ]]; then
+    echo "PULSE: Cold (Zero files)"
+  else
+    echo "PULSE: Warm ($files files)"
+  fi
+}
+
+if [[ "${1:-}" == "--pulse" ]]; then
+  check_pulse
+  exit 0
+fi
+
 ensure_reposcry_gitignore
 refuse_tracked_reposcry_cache
 

package/scripts/reposcry-task-context.sh

@@ -20,7 +20,15 @@ cd "$REPO_ROOT"
 "$SCRIPT_DIR/reposcry-bootstrap.sh" || true
 
 mkdir -p .reposcry
-BUDGET=${REPOSCRY_CONTEXT_BUDGET:-20000}
+
+# Detect low-token modes (Caveman, etc.)
+DEFAULT_BUDGET=20000
+if [[ "${CAVEMAN_MODE:-off}" != "off" ]] || [[ "${LOW_TOKEN_MODE:-0}" == "1" ]]; then
+  DEFAULT_BUDGET=5000
+  echo "Low-token mode detected: reducing context budget to $DEFAULT_BUDGET"
+fi
+
+BUDGET=${REPOSCRY_CONTEXT_BUDGET:-$DEFAULT_BUDGET}
 OUTPUT_PATH=".reposcry/AI_CONTEXT.md"
 
 echo "Generating RepoScry task context at $OUTPUT_PATH..."

package/triage/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: triage
-description: "Ingest or resume a tracked/local task in the RPIV workspace. Use when starting or returning to jira:, github:, gitlab:, or local: work and you need canonical WORK.md state without duplicating scaffold."
+description: "Ingest or resume a tracked/local task in the RPIV workspace. Use when starting or returning to jira:, github:, gitlab:, or local: work and you need canonical WORK.md state without duplicating the scaffold."
 ---
 
 # Skill: triage
@@ -8,10 +8,10 @@ description: "Ingest or resume a tracked/local task in the RPIV workspace. Use w
 Start RPIV by creating, resuming, or explicitly reopening a task workspace.
 
 ## Guardrails
-- READ: user argument, `.workflow/active_task.json` when present, target `metadata.json`, and target `WORK.md` when resuming.
-- WRITE: `.workflow/tasks/[source-id]/WORK.md`, `.workflow/tasks/[source-id]/metadata.json`, `.workflow/active_task.json`; optional `.reposcry/` cache files only when RepoScry is installed.
+- READ: user argument, `.workflow/active_task.json` if present, target `metadata.json`, and target `WORK.md` if resuming.
+- WRITE: `.workflow/tasks/[source-id]/WORK.md`, `.workflow/tasks/[source-id]/metadata.json`, `.workflow/active_task.json`; optional `.reposcry/` cache files only if RepoScry is installed.
 - On **create**, initialize required guarded sections only if absent: `[BRIEF]`, `[GRILL]`, `[PLAN]`, `[LOG]`, `[META]`.
-- On **resume**, only update `.workflow/active_task.json`, metadata timestamps/state as needed, `[META]`, and append one concise `[LOG]` entry.
+- On **resume**, only update `.workflow/active_task.json`, metadata timestamps/state as needed, and `[META]`, then append one concise `[LOG]` entry.
 - NEVER: duplicate guarded sections.
 - NEVER: overwrite existing `[BRIEF]`, `[GRILL]`, or `[PLAN]` during triage.
 - NEVER: create `PROBLEM.md`, `PRD.md`, `PLAN.md`, or `EVIDENCE.md`.
@@ -21,49 +21,34 @@ Start RPIV by creating, resuming, or explicitly reopening a task workspace.
 - NEVER: mutate `done` or `archived` tasks unless the user explicitly requested `reopen`, `fresh`, or `reset`.
 
 ## Command forms
-- `/triage local:setup-v2` — auto mode: create if missing, resume if active/blocked.
-- `/triage resume local:setup-v2` — resume existing task; create if missing only when helper reports that behavior.
-- `/triage reopen local:setup-v2` — explicitly reopen a done/archived task and mark it active.
-- `/triage fresh local:setup-v2` — archive/reset the previous task workspace and create a fresh one.
-- `/triage reset local:setup-v2` — alias for fresh/reset behavior when supported by the helper.
 
-## Semantics
-
-### Create
-Use when the target task folder does not exist.
-- Fetch remote issue data for `github:`, `gitlab:`, or `jira:` sources.
-- Create `.workflow/tasks/[source-id]/metadata.json` with at least `id`, `source`, `taskFolder`, `branch`, `status`, `phase`, `createdAt`, `updatedAt`.
-- Create `.workflow/tasks/[source-id]/WORK.md` with guarded RPIV sections.
-- Set `status=active` and `phase=triaged`.
-- Write canonical `.workflow/active_task.json`.
-
-### Resume
-Use when the task exists and `status` is `active`, `blocked`, or missing/legacy.
-- Do not refetch remote data.
-- Do not rewrite the brief, grill notes, or plan.
+### /triage local:setup-v2
+- **Standard/auto mode**: create if missing, resume if active/blocked.
+- Will not rewrite brief, grill notes, or plan.
 - Backfill missing metadata fields without destroying unknown fields.
 - Refresh active pointer and human-readable `[META]`.
 - Append one timestamped `[LOG]` entry such as `Task resumed via /triage`.
 
-### Reopen
-Use only when the user explicitly asks to reopen a task that is `done` or `archived`.
+### /triage local:setup-v2 reopen
+- Use only if user explicitly asks to reopen a task that is `done` or `archived`.
 - Mark `status=active`.
 - Preserve existing `WORK.md` history.
 - Append a `[LOG]` entry explaining the reopen.
 - Recommend returning to the next valid RPIV stage based on existing sections.
 
-### Fresh / reset
-Use only when the user explicitly asks to start over.
-- Preserve or archive the old task workspace before creating a new one when the helper supports it.
-- Create a new task workspace with clean guarded sections.
+### /triage local:setup-v2 fresh / reset
+- Use only if user explicitly asks to start over.
+- Preserve or archive old task workspace before creating new one when helper supports it.
+- Create new task workspace with clean guarded sections.
 - Do not silently delete task history.
 
 ### Refuse
-Use when auto/resume mode targets a `done` or `archived` task.
+- Use when auto/resume mode targets a `done` or `archived` task.
 - Stop instead of mutating.
-- Tell the user to choose `reopen` or `fresh/reset`.
+- Tell user to choose `reopen` or `fresh/reset`.
 
 ## Workflow
+
 1. Parse optional mode and required namespace:
    - `jira:PROJ-123` -> `jira-PROJ-123`
    - `github:42` -> `github-42`
@@ -72,23 +57,24 @@ Use when auto/resume mode targets a `done` or `archived` task.
 2. Run the bundled helper script: `../scripts/triage_helper.sh <source> <id> [mode]`.
    - Preferred absolute path: `<skill_dir>/../scripts/triage_helper.sh`.
 3. Read the resulting active pointer, metadata, and `WORK.md`.
-4. Determine action from helper output and task state: `created`, `resumed`, `reopened`, `refused`, or `fresh/reset`.
-5. Optional RepoScry bootstrap: if the bundled `../scripts/reposcry-bootstrap.sh` is available, run it to seed `.reposcry/` for later `/frame` and `/grill-with-docs`. The helper must ensure `.reposcry/` is ignored, must stop if `.reposcry/` is tracked/staged, and should continue normally when RepoScry is unavailable.
-6. **Technical Pre-check (Repo Pulse)** for created/reopened/resumed tasks:
-   - Extract key filenames, classes, commands, or keywords from the issue/task description.
-   - Verify existence on the current branch when concrete files are named.
-   - Check recent commits or open PRs only when relevant to the named files/scope.
-   - Version-check environment/dependencies only when the task depends on a version claim.
-7. Classify the task: Problem / Proposal.
-8. **Log Findings**: append concise Repo Pulse and classification findings to `[LOG]` only. Do not edit `[BRIEF]`, `[GRILL]`, or `[PLAN]`.
-9. **Branch Guidance**: record the current branch in `[META]`. For `jira:` tasks, also preserve the Jira key in `[META]` and mention that `/implement` must use it in the commit subject. Planning on `main`/`master` is allowed; implementation must use `/implement` branch enforcement.
-10. End by recommending the next valid command:
-   - newly created -> `/frame`
-   - existing with empty `[BRIEF]` -> `/frame`
-   - framed but not grilled -> `/grill-with-docs`
-   - grilled but not planned -> `/plan`
-   - planned -> await explicit `/implement` or `EXECUTE`
-   - refused -> ask user to choose `reopen` or `fresh/reset`
+4. Determine action based on helper output and task state: `created`, `resumed`, `reopened`, `refused`, `fresh/reset`.
+5. Optional RepoScry bootstrap: if bundled `../scripts/reposcry-bootstrap.sh` available, run it to seed `.reposcry/` for later `/frame` and `/grill-with-docs`.
+   - Run `../scripts/reposcry-bootstrap.sh --pulse` to determine "Repo Pulse" (Warm/Cold/Missing) for the output contract.
+   - Run `../scripts/reposcry-bootstrap.sh` to ensure the local cache is initialized.
+   - The helper must ensure `.reposcry/` is ignored, must stop if `.reposcry/` tracked/staged, and continue normally if RepoScry is unavailable.
+6. Verify branch: ensure current branch matches metadata `branch` (or is `main`/`master` for planning).
+7. If task was newly created, infer classification:
+   - `github:`, `gitlab:`, `jira:` labels/type often indicate Problem (bug) or Proposal (feature).
+   - Local tasks default to Proposal unless specified.
+8. If resuming, check `[BRIEF]` and `[PLAN]` completion status.
+9. **Final Guidance**: record current branch in `[META]`. For `jira:` tasks, also preserve the Jira key in `[META]` and mention `/implement` must use it in the commit subject. Planning on `main`/`master` allowed; implementation will use `/implement` branch enforcement.
+10. End by recommending next valid command:
+    - newly created -> `/frame`
+    - existing empty `[BRIEF]` -> `/frame`
+    - framed but not grilled -> `/grill-with-docs`
+    - grilled but not planned -> `/plan`
+    - planned -> await explicit `/implement` or `EXECUTE`
+    - refused -> ask user to choose `reopen` or `fresh/reset`
 
 ## Output contract
 End with:
@@ -97,6 +83,6 @@ End with:
 - **Status**: active / blocked / done / archived / unknown
 - **Phase**: triaged / framed / grilled / planned / implementing / verifying / synced / closed / unknown
 - **Branch**: current branch
-- **Repo Pulse**: Found / Missing / Outdated / Not applicable
+- **Repo Pulse**: Warm / Cold / Missing / Not applicable
 - **Classification**: Problem / Proposal
 - **Next step**: `/frame`, `/grill-with-docs`, `/plan`, `/implement`, or explicit reopen/fresh choice

package/verify/SKILL.md

@@ -19,7 +19,7 @@ The final gate for a slice or task. Verify truth before reporting progress.
 2. Run stated verification commands and available quality gates.
 3. If RepoScry is available, optionally add graph-aware evidence with commands such as `reposcry validate main HEAD` and `reposcry --repo . get_affected_flows main HEAD`. Treat RepoScry as supplemental evidence, not a hard requirement. Verify `.reposcry/` is not staged or tracked before reporting review readiness.
 4. Check for AI artifacts: placeholder comments, fake APIs, dead code, inconsistent naming.
-5. Confirm commit messages include Conventional Commit format and `Assisted-by: [AGENT]:[MODEL] [tools]` when AI contributed.
+5. Confirm commit messages include Conventional Commit format. When AI contributed, you MUST append an `Assisted-by` trailer. Do NOT guess or hallucinate the model name from your system prompt. You must run `bash ~/RQZ/personal/nothing/packages/norpiv/scripts/get-pi-model.sh` and use its exact output to construct the trailer: `Assisted-by: <EXACT_OUTPUT> [tools]`.
 6. If passing, mark the slice checkbox complete in `[PLAN]` and append verification evidence to `[LOG]` (Format: `YYYY-MM-DD hh:mm AM/PM`).
 7. Recommend `/sync` for tracker update, or cleanup if the task is fully merged and user approves.