@nforma.ai/nforma 0.2.1 → 0.29.0

package/README.md

@@ -1,8 +1,8 @@
 <div align="center">
 
-# nForma — Quorum Gets Shit Done
+# nForma — Consensus before code. Proof before production.
 
-**Every planning decision verified by a quorum of AI models before Claude executes a single line.**
+**nForma orchestrates a quorum of diverse coding agents that debate a plan until they reach full consensus.<br>Then formal methods turn requirements into invariants and simulate the system ahead of time.<br>The result: fewer hallucinations, fewer blind spots, and systems designed to survive the future.**
 
 [![npm version](https://img.shields.io/npm/v/@nforma.ai/nforma?style=for-the-badge&logo=npm&logoColor=white&color=CB3837)](https://www.npmjs.com/package/@nforma.ai/nforma)
 [![npm downloads](https://img.shields.io/npm/dm/@nforma.ai/nforma?style=for-the-badge&logo=npm&logoColor=white&color=CB3837)](https://www.npmjs.com/package/@nforma.ai/nforma)

package/agents/{qgsd-codebase-mapper.md → nf-codebase-mapper.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-codebase-mapper
+name: nf-codebase-mapper
 description: Explores codebase and writes structured analysis documents. Spawned by map-codebase with a focus area (tech, arch, quality, concerns). Writes documents directly to reduce orchestrator context load.
 tools: Read, Bash, Grep, Glob, Write
 color: cyan

package/agents/{qgsd-debugger.md → nf-debugger.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-debugger
+name: nf-debugger
 description: Investigates bugs using scientific method, manages debug sessions, handles checkpoints. Spawned by /qgsd:debug orchestrator.
 tools: Read, Write, Edit, Bash, Grep, Glob, WebSearch
 color: orange
@@ -985,7 +985,7 @@ mv .planning/debug/{slug}.md .planning/debug/resolved/
 **Check planning config using state load (commit_docs is available from the output):**
 
 ```bash
-INIT=$(node ~/.claude/qgsd/bin/gsd-tools.cjs state load)
+INIT=$(node ~/.claude/nf/bin/gsd-tools.cjs state load)
 # commit_docs is in the JSON output
 ```
 
@@ -1002,7 +1002,7 @@ Root cause: {root_cause}"
 
 Then commit planning docs via CLI (respects `commit_docs` config automatically):
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs commit "docs: resolve debug {slug}" --files .planning/debug/resolved/{slug}.md
+node ~/.claude/nf/bin/gsd-tools.cjs commit "docs: resolve debug {slug}" --files .planning/debug/resolved/{slug}.md
 ```
 
 Report completion and offer next steps.

package/agents/{qgsd-executor.md → nf-executor.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-executor
+name: nf-executor
 description: Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.
 tools: Read, Write, Edit, Bash, Grep, Glob
 color: yellow
@@ -37,7 +37,7 @@ This ensures project-specific patterns, conventions, and best practices are appl
 Load execution context:
 
 ```bash
-INIT=$(node ~/.claude/qgsd/bin/gsd-tools.cjs init execute-phase "${PHASE}")
+INIT=$(node ~/.claude/nf/bin/gsd-tools.cjs init execute-phase "${PHASE}")
 ```
 
 Extract from init JSON: `executor_model`, `commit_docs`, `phase_dir`, `plans`, `incomplete_plans`.
@@ -193,7 +193,7 @@ Track auto-fix attempts per task. After 3 auto-fix attempts on a single task:
 Check if auto mode is active at executor start:
 
 ```bash
-AUTO_CFG=$(node ~/.claude/qgsd/bin/gsd-tools.cjs config-get workflow.auto_advance 2>/dev/null || echo "false")
+AUTO_CFG=$(node ~/.claude/nf/bin/gsd-tools.cjs config-get workflow.auto_advance 2>/dev/null || echo "false")
 ```
 
 Store the result for checkpoint handling below.
@@ -206,7 +206,7 @@ Store the result for checkpoint handling below.
 Before any `checkpoint:human-verify`, ensure verification environment is ready. If plan lacks server startup before checkpoint, ADD ONE (deviation Rule 3).
 
 For full automation-first patterns, server lifecycle, CLI handling:
-**See @~/.claude/qgsd/references/checkpoints.md**
+**See @~/.claude/nf/references/checkpoints.md**
 
 **Quick reference:** Users NEVER run CLI commands. Users ONLY visit URLs, click UI, evaluate visuals, provide secrets. Claude does all automation.
 
@@ -329,7 +329,7 @@ After all tasks complete, create `{phase}-{plan}-SUMMARY.md` at `.planning/phase
 
 **ALWAYS use the Write tool to create files** — never use `Bash(cat << 'EOF')` or heredoc commands for file creation.
 
-**Use template:** @~/.claude/qgsd/templates/summary.md
+**Use template:** @~/.claude/nf/templates/summary.md
 
 **Frontmatter:** phase, plan, subsystem, tags, dependency graph (requires/provides/affects), tech-stack (added/patterns), key-files (created/modified), decisions, metrics (duration, completed date).
 
@@ -382,34 +382,34 @@ After SUMMARY.md, update STATE.md using gsd-tools:
 
 ```bash
 # Advance plan counter (handles edge cases automatically)
-node ~/.claude/qgsd/bin/gsd-tools.cjs state advance-plan
+node ~/.claude/nf/bin/gsd-tools.cjs state advance-plan
 
 # Recalculate progress bar from disk state
-node ~/.claude/qgsd/bin/gsd-tools.cjs state update-progress
+node ~/.claude/nf/bin/gsd-tools.cjs state update-progress
 
 # Record execution metrics
-node ~/.claude/qgsd/bin/gsd-tools.cjs state record-metric \
+node ~/.claude/nf/bin/gsd-tools.cjs state record-metric \
   --phase "${PHASE}" --plan "${PLAN}" --duration "${DURATION}" \
   --tasks "${TASK_COUNT}" --files "${FILE_COUNT}"
 
 # Add decisions (extract from SUMMARY.md key-decisions)
 for decision in "${DECISIONS[@]}"; do
-  node ~/.claude/qgsd/bin/gsd-tools.cjs state add-decision \
+  node ~/.claude/nf/bin/gsd-tools.cjs state add-decision \
     --phase "${PHASE}" --summary "${decision}"
 done
 
 # Update session info
-node ~/.claude/qgsd/bin/gsd-tools.cjs state record-session \
+node ~/.claude/nf/bin/gsd-tools.cjs state record-session \
   --stopped-at "Completed ${PHASE}-${PLAN}-PLAN.md"
 ```
 
 ```bash
 # Update ROADMAP.md progress for this phase (plan counts, status)
-node ~/.claude/qgsd/bin/gsd-tools.cjs roadmap update-plan-progress "${PHASE_NUMBER}"
+node ~/.claude/nf/bin/gsd-tools.cjs roadmap update-plan-progress "${PHASE_NUMBER}"
 
 # Mark completed requirements from PLAN.md frontmatter
 # Extract the `requirements` array from the plan's frontmatter, then mark each complete
-node ~/.claude/qgsd/bin/gsd-tools.cjs requirements mark-complete ${REQ_IDS}
+node ~/.claude/nf/bin/gsd-tools.cjs requirements mark-complete ${REQ_IDS}
 ```
 
 **Requirement IDs:** Extract from the PLAN.md frontmatter `requirements:` field (e.g., `requirements: [AUTH-01, AUTH-02]`). Pass all IDs to `requirements mark-complete`. If the plan has no requirements field, skip this step.
@@ -427,13 +427,13 @@ node ~/.claude/qgsd/bin/gsd-tools.cjs requirements mark-complete ${REQ_IDS}
 
 **For blockers found during execution:**
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs state add-blocker "Blocker description"
+node ~/.claude/nf/bin/gsd-tools.cjs state add-blocker "Blocker description"
 ```
 </state_updates>
 
 <final_commit>
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs commit "docs({phase}-{plan}): complete [plan-name] plan" --files .planning/phases/XX-name/{phase}-{plan}-SUMMARY.md .planning/STATE.md .planning/ROADMAP.md .planning/REQUIREMENTS.md
+node ~/.claude/nf/bin/gsd-tools.cjs commit "docs({phase}-{plan}): complete [plan-name] plan" --files .planning/phases/XX-name/{phase}-{plan}-SUMMARY.md .planning/STATE.md .planning/ROADMAP.md .planning/REQUIREMENTS.md
 ```
 
 Separate from per-task commits — captures execution results only.

package/agents/{qgsd-integration-checker.md → nf-integration-checker.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-integration-checker
+name: nf-integration-checker
 description: Verifies cross-phase integration and E2E flows. Checks that phases connect properly and user workflows complete end-to-end.
 tools: Read, Bash, Grep, Glob
 color: blue

package/agents/{qgsd-phase-researcher.md → nf-phase-researcher.md}

@@ -1,6 +1,6 @@
 ---
-name: qgsd-phase-researcher
-description: Researches how to implement a phase before planning. Produces RESEARCH.md consumed by qgsd-planner. Spawned by /qgsd:plan-phase orchestrator.
+name: nf-phase-researcher
+description: Researches how to implement a phase before planning. Produces RESEARCH.md consumed by nf-planner. Spawned by /qgsd:plan-phase orchestrator.
 tools: Read, Write, Bash, Grep, Glob, WebSearch, WebFetch, mcp__context7__*
 color: cyan
 ---
@@ -49,7 +49,7 @@ If CONTEXT.md exists, it constrains your research scope. Don't explore alternati
 </upstream_input>
 
 <downstream_consumer>
-Your RESEARCH.md is consumed by `qgsd-planner`:
+Your RESEARCH.md is consumed by `nf-planner`:
 
 | Section | How Planner Uses It |
 |---------|---------------------|
@@ -120,7 +120,7 @@ When researching "best library for X": find what the ecosystem actually uses, do
 Check `brave_search` from init context. If `true`, use Brave Search for higher quality results:
 
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs websearch "your query" --limit 10
+node ~/.claude/nf/bin/gsd-tools.cjs websearch "your query" --limit 10
 ```
 
 **Options:**
@@ -330,7 +330,7 @@ Orchestrator provides: phase number/name, description/goal, requirements, constr
 
 Load phase context using init command:
 ```bash
-INIT=$(node ~/.claude/qgsd/bin/gsd-tools.cjs init phase-op "${PHASE}")
+INIT=$(node ~/.claude/nf/bin/gsd-tools.cjs init phase-op "${PHASE}")
 ```
 
 Extract from init JSON: `phase_dir`, `padded_phase`, `phase_number`, `commit_docs`.
@@ -417,7 +417,7 @@ Write to: `$PHASE_DIR/$PADDED_PHASE-RESEARCH.md`
 ## Step 6: Commit Research (optional)
 
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs commit "docs($PHASE): research phase domain" --files "$PHASE_DIR/$PADDED_PHASE-RESEARCH.md"
+node ~/.claude/nf/bin/gsd-tools.cjs commit "docs($PHASE): research phase domain" --files "$PHASE_DIR/$PADDED_PHASE-RESEARCH.md"
 ```
 
 ## Step 7: Return Structured Result

package/agents/{qgsd-plan-checker.md → nf-plan-checker.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-plan-checker
+name: nf-plan-checker
 description: Verifies plans will achieve phase goal before execution. Goal-backward analysis of plan quality. Spawned by /qgsd:plan-phase orchestrator.
 tools: Read, Bash, Glob, Grep
 color: green
@@ -72,8 +72,8 @@ Goal-backward verification works backwards from outcome:
 Then verify each level against the actual plan files.
 
 **The difference:**
-- `qgsd-verifier`: Verifies code DID achieve goal (after execution)
-- `qgsd-plan-checker`: Verifies plans WILL achieve goal (before execution)
+- `nf-verifier`: Verifies code DID achieve goal (after execution)
+- `nf-plan-checker`: Verifies plans WILL achieve goal (before execution)
 
 Same methodology (goal-backward), different timing, different subject matter.
 </core_principle>
@@ -320,7 +320,7 @@ issue:
 
 Load phase operation context:
 ```bash
-INIT=$(node ~/.claude/qgsd/bin/gsd-tools.cjs init phase-op "${PHASE_ARG}")
+INIT=$(node ~/.claude/nf/bin/gsd-tools.cjs init phase-op "${PHASE_ARG}")
 ```
 
 Extract from init JSON: `phase_dir`, `phase_number`, `has_plans`, `plan_count`.
@@ -329,7 +329,7 @@ Orchestrator provides CONTEXT.md content in the verification prompt. If provided
 
 ```bash
 ls "$phase_dir"/*-PLAN.md 2>/dev/null
-node ~/.claude/qgsd/bin/gsd-tools.cjs roadmap get-phase "$phase_number"
+node ~/.claude/nf/bin/gsd-tools.cjs roadmap get-phase "$phase_number"
 ls "$phase_dir"/*-BRIEF.md 2>/dev/null
 ```
 
@@ -342,7 +342,7 @@ Use gsd-tools to validate plan structure:
 ```bash
 for plan in "$PHASE_DIR"/*-PLAN.md; do
   echo "=== $plan ==="
-  PLAN_STRUCTURE=$(node ~/.claude/qgsd/bin/gsd-tools.cjs verify plan-structure "$plan")
+  PLAN_STRUCTURE=$(node ~/.claude/nf/bin/gsd-tools.cjs verify plan-structure "$plan")
   echo "$PLAN_STRUCTURE"
 done
 ```
@@ -360,7 +360,7 @@ Map errors/warnings to verification dimensions:
 Extract must_haves from each plan using gsd-tools:
 
 ```bash
-MUST_HAVES=$(node ~/.claude/qgsd/bin/gsd-tools.cjs frontmatter get "$PLAN_PATH" --field must_haves)
+MUST_HAVES=$(node ~/.claude/nf/bin/gsd-tools.cjs frontmatter get "$PLAN_PATH" --field must_haves)
 ```
 
 Returns JSON: `{ truths: [...], artifacts: [...], key_links: [...] }`
@@ -403,7 +403,7 @@ For each requirement: find covering task(s), verify action is specific, flag gap
 Use gsd-tools plan-structure verification (already run in Step 2):
 
 ```bash
-PLAN_STRUCTURE=$(node ~/.claude/qgsd/bin/gsd-tools.cjs verify plan-structure "$PLAN_PATH")
+PLAN_STRUCTURE=$(node ~/.claude/nf/bin/gsd-tools.cjs verify plan-structure "$PLAN_PATH")
 ```
 
 The `tasks` array in the result shows each task's completeness:
@@ -603,7 +603,7 @@ Plans verified. Run `/qgsd:execute-phase {phase}` to proceed.
 
 <anti_patterns>
 
-**DO NOT** check code existence — that's qgsd-verifier's job. You verify plans, not codebase.
+**DO NOT** check code existence — that's nf-verifier's job. You verify plans, not codebase.
 
 **DO NOT** run the application. Static plan analysis only.
 

package/agents/{qgsd-planner.md → nf-planner.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-planner
+name: nf-planner
 description: Creates executable phase plans with task breakdown, dependency analysis, and goal-backward verification. Spawned by /qgsd:plan-phase orchestrator.
 tools: Read, Write, Bash, Glob, Grep, WebFetch, mcp__context7__*
 color: green
@@ -380,8 +380,8 @@ Output: [Artifacts created]
 </objective>
 
 <execution_context>
-@~/.claude/qgsd/workflows/execute-plan.md
-@~/.claude/qgsd/templates/summary.md
+@~/.claude/nf/workflows/execute-plan.md
+@~/.claude/nf/templates/summary.md
 </execution_context>
 
 <context>
@@ -817,7 +817,7 @@ Group by plan, dimension, severity.
 ### Step 6: Commit
 
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs commit "fix($PHASE): revise plans based on checker feedback" --files .planning/phases/$PHASE-*/$PHASE-*-PLAN.md
+node ~/.claude/nf/bin/gsd-tools.cjs commit "fix($PHASE): revise plans based on checker feedback" --files .planning/phases/$PHASE-*/$PHASE-*-PLAN.md
 ```
 
 ### Step 7: Return Revision Summary
@@ -856,7 +856,7 @@ node ~/.claude/qgsd/bin/gsd-tools.cjs commit "fix($PHASE): revise plans based on
 Load planning context:
 
 ```bash
-INIT=$(node ~/.claude/qgsd/bin/gsd-tools.cjs init plan-phase "${PHASE}")
+INIT=$(node ~/.claude/nf/bin/gsd-tools.cjs init plan-phase "${PHASE}")
 ```
 
 Extract from init JSON: `planner_model`, `researcher_model`, `checker_model`, `commit_docs`, `research_enabled`, `phase_dir`, `phase_number`, `has_research`, `has_context`.
@@ -912,7 +912,7 @@ Apply discovery level protocol (see discovery_levels section).
 
 **Step 1 — Generate digest index:**
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs history-digest
+node ~/.claude/nf/bin/gsd-tools.cjs history-digest
 ```
 
 **Step 2 — Select relevant phases (typically 2-4):**
@@ -1030,7 +1030,7 @@ Include all frontmatter fields.
 Validate each created PLAN.md using gsd-tools:
 
 ```bash
-VALID=$(node ~/.claude/qgsd/bin/gsd-tools.cjs frontmatter validate "$PLAN_PATH" --schema plan)
+VALID=$(node ~/.claude/nf/bin/gsd-tools.cjs frontmatter validate "$PLAN_PATH" --schema plan)
 ```
 
 Returns JSON: `{ valid, missing, present, schema }`
@@ -1043,7 +1043,7 @@ Required plan frontmatter fields:
 Also validate plan structure:
 
 ```bash
-STRUCTURE=$(node ~/.claude/qgsd/bin/gsd-tools.cjs verify plan-structure "$PLAN_PATH")
+STRUCTURE=$(node ~/.claude/nf/bin/gsd-tools.cjs verify plan-structure "$PLAN_PATH")
 ```
 
 Returns JSON: `{ valid, errors, warnings, task_count, tasks }`
@@ -1080,7 +1080,7 @@ Plans:
 
 <step name="git_commit">
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs commit "docs($PHASE): create phase plan" --files .planning/phases/$PHASE-*/$PHASE-*-PLAN.md .planning/ROADMAP.md
+node ~/.claude/nf/bin/gsd-tools.cjs commit "docs($PHASE): create phase plan" --files .planning/phases/$PHASE-*/$PHASE-*-PLAN.md .planning/ROADMAP.md
 ```
 </step>
 

package/agents/{qgsd-project-researcher.md → nf-project-researcher.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-project-researcher
+name: nf-project-researcher
 description: Researches domain ecosystem before roadmap creation. Produces files in .planning/research/ consumed during roadmap creation. Spawned by /qgsd:new-project or /qgsd:new-milestone orchestrators.
 tools: Read, Write, Bash, Grep, Glob, WebSearch, WebFetch, mcp__context7__*
 color: cyan
@@ -99,7 +99,7 @@ Always include current year. Use multiple query variations. Mark WebSearch-only
 Check `brave_search` from orchestrator context. If `true`, use Brave Search for higher quality results:
 
 ```bash
-node ~/.claude/qgsd/bin/gsd-tools.cjs websearch "your query" --limit 10
+node ~/.claude/nf/bin/gsd-tools.cjs websearch "your query" --limit 10
 ```
 
 **Options:**

package/agents/{qgsd-quorum-orchestrator.md → nf-quorum-orchestrator.md}

@@ -1,15 +1,15 @@
-<!-- DEPRECATED: This agent is superseded by direct inline dispatch in commands/qgsd/quorum.md as of quick-103. The orchestrator Task-spawn indirection is no longer needed — quorum.md now contains the full R3 protocol inline (with qgsd-quorum-slot-worker for per-slot dispatch). Retained for reference only. Do not spawn this agent. -->
+<!-- DEPRECATED: This agent is superseded by direct inline dispatch in commands/nf/quorum.md as of quick-103. The orchestrator Task-spawn indirection is no longer needed — quorum.md now contains the full R3 protocol inline (with nf-quorum-slot-worker for per-slot dispatch). Retained for reference only. Do not spawn this agent. -->
 ---
-name: qgsd-quorum-orchestrator
+name: nf-quorum-orchestrator
 description: >
   DEPRECATED — do not spawn. Superseded by direct inline dispatch in
-  commands/qgsd/quorum.md as of quick-103. Retained for reference only.
+  commands/nf/quorum.md as of quick-103. Retained for reference only.
 tools: Read, Write, Bash, Task, Glob, Grep
 color: purple
 ---
 
 <role>
-You are the QGSD quorum orchestrator. When invoked, execute the full R3 quorum protocol
+You are the nForma quorum orchestrator. When invoked, execute the full R3 quorum protocol
 for the question or bundle passed in `$ARGUMENTS`.
 
 **SEQUENTIAL CALLS ONLY — NO SIBLING TOOL CALLS.**
@@ -55,7 +55,7 @@ These two variables (`$ARTIFACT_PATH`, `$REPO_DIR`) are available to all subsequ
 Run before any model calls:
 
 ```bash
-node "$HOME/.claude/qgsd-bin/check-provider-health.cjs" --json
+node "$HOME/.claude/nf-bin/check-provider-health.cjs" --json
 ```
 
 Parse the JSON output. Build:
@@ -66,12 +66,12 @@ Parse the JSON output. Build:
 
 Any server with `available: false` → mark UNAVAIL immediately — skip inference calls entirely.
 
-3. **`$QUORUM_ACTIVE`**: read from `~/.claude/qgsd.json` (project config takes precedence):
+3. **`$QUORUM_ACTIVE`**: read from `~/.claude/nf.json` (project config takes precedence):
 ```bash
 node -e "
 const fs = require('fs'), os = require('os'), path = require('path');
-const globalCfg = path.join(os.homedir(), '.claude', 'qgsd.json');
-const projCfg   = path.join(process.cwd(), '.claude', 'qgsd.json');
+const globalCfg = path.join(os.homedir(), '.claude', 'nf.json');
+const projCfg   = path.join(process.cwd(), '.claude', 'nf.json');
 let cfg = {};
 for (const f of [globalCfg, projCfg]) {
   try { Object.assign(cfg, JSON.parse(fs.readFileSync(f, 'utf8'))); } catch(_){}
@@ -87,12 +87,12 @@ A server in `$QUORUM_ACTIVE` but absent from `$CLAUDE_MCP_SERVERS` = skip silent
 - For each server with `available: false`, log: `Pre-flight skip: <serverName> (<providerName> DOWN)`
 - Remove these from the working list for all subsequent steps.
 
-Read `preferSub` and `agent_config` from qgsd.json (project config takes precedence):
+Read `preferSub` and `agent_config` from nf.json (project config takes precedence):
 ```bash
 node -e "
 const fs = require('fs'), os = require('os'), path = require('path');
-const globalCfg = path.join(os.homedir(), '.claude', 'qgsd.json');
-const projCfg   = path.join(process.cwd(), '.claude', 'qgsd.json');
+const globalCfg = path.join(os.homedir(), '.claude', 'nf.json');
+const projCfg   = path.join(process.cwd(), '.claude', 'nf.json');
 let cfg = {};
 for (const f of [globalCfg, projCfg]) {
   try { Object.assign(cfg, JSON.parse(fs.readFileSync(f, 'utf8'))); } catch(_) {}
@@ -109,12 +109,12 @@ Store result as `$PREFER_SUB_CONFIG`.
 - **Shuffle within groups:** Shuffle the sub group and the api group independently. Final order when preferSub=true: shuffle(sub), shuffle(api). When preferSub=false: shuffle all slots.
 - Log: `Active slots: <slot1>, <slot2>, ...`
 
-**min_quorum_size check:** Read from `~/.claude/qgsd.json` (project config takes precedence; default: 3 if absent):
+**min_quorum_size check:** Read from `~/.claude/nf.json` (project config takes precedence; default: 3 if absent):
 ```bash
 node -e "
 const fs = require('fs'), os = require('os'), path = require('path');
-const globalCfg = path.join(os.homedir(), '.claude', 'qgsd.json');
-const projCfg   = path.join(process.cwd(), '.claude', 'qgsd.json');
+const globalCfg = path.join(os.homedir(), '.claude', 'nf.json');
+const projCfg   = path.join(process.cwd(), '.claude', 'nf.json');
 let cfg = {};
 for (const f of [globalCfg, projCfg]) {
   try { Object.assign(cfg, JSON.parse(fs.readFileSync(f, 'utf8'))); } catch(_){}
@@ -136,7 +136,7 @@ If `availableCount < min_quorum_size`:
 **Availability cache check:** After building the provider slot list but before the min_quorum_size check, read the scoreboard availability data:
 
 ```bash
-node "$HOME/.claude/qgsd-bin/update-scoreboard.cjs" get-availability \
+node "$HOME/.claude/nf-bin/update-scoreboard.cjs" get-availability \
   --scoreboard .planning/quorum-scoreboard.json 2>/dev/null || echo '{}'
 ```
 
@@ -171,7 +171,7 @@ const fs = require('fs'), path = require('path'), os = require('os');
 
 // Find providers.json
 const searchPaths = [
-  path.join(os.homedir(), '.claude', 'qgsd-bin', 'providers.json'),
+  path.join(os.homedir(), '.claude', 'nf-bin', 'providers.json'),
 ];
 try {
   const cj = JSON.parse(fs.readFileSync(path.join(os.homedir(), '.claude.json'), 'utf8'));
@@ -186,8 +186,8 @@ for (const p of searchPaths) {
 }
 
 // Read quorum_active
-const globalCfg = path.join(os.homedir(), '.claude', 'qgsd.json');
-const projCfg   = path.join(process.cwd(), '.claude', 'qgsd.json');
+const globalCfg = path.join(os.homedir(), '.claude', 'nf.json');
+const projCfg   = path.join(process.cwd(), '.claude', 'nf.json');
 let cfg = {};
 for (const f of [globalCfg, projCfg]) {
   try { Object.assign(cfg, JSON.parse(fs.readFileSync(f, 'utf8'))); } catch(_) {}
@@ -206,7 +206,7 @@ console.log(JSON.stringify(team));
 Detect Claude model ID: `CLAUDE_MODEL` env → `ANTHROPIC_MODEL` env → session model from context.
 
 ```bash
-node "$HOME/.claude/qgsd-bin/update-scoreboard.cjs" init-team \
+node "$HOME/.claude/nf-bin/update-scoreboard.cjs" init-team \
   --claude-model "<claude_model_id>" \
   --team '<TEAM_JSON>'
 ```
@@ -220,12 +220,12 @@ node "$HOME/.claude/qgsd-bin/update-scoreboard.cjs" init-team \
 ### Parse question
 
 The question is `$ARGUMENTS`. If empty or too short, stop with:
-`"No question provided. Pass question as: Task(subagent_type=qgsd-quorum-orchestrator, prompt='<question>')"`
+`"No question provided. Pass question as: Task(subagent_type=nf-quorum-orchestrator, prompt='<question>')"`
 
 Display:
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- QGSD ► QUORUM: Mode A — Pure Question
+ nForma ► QUORUM: Mode A — Pure Question
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Question: [question]
@@ -256,7 +256,7 @@ $ALL_ROUND_RESULTS = []   (accumulates results across all rounds for scoreboard)
 Display before dispatching each round:
 ```
 ─────────────────────────────────────────────
- QGSD ► QUORUM Round $CURRENT_ROUND / up to $MAX_ROUNDS
+ nForma ► QUORUM Round $CURRENT_ROUND / up to $MAX_ROUNDS
 ─────────────────────────────────────────────
 ```
 
@@ -266,7 +266,7 @@ All workers for this round are dispatched as parallel sibling Task calls in one
 
 ```
 Task(
-  subagent_type="qgsd-quorum-slot-worker",
+  subagent_type="nf-quorum-slot-worker",
   description="<slotName> quorum R<$CURRENT_ROUND>",
   prompt="""
 slot: <slotName>
@@ -290,7 +290,7 @@ Collect all worker result blocks → store as `$ROUND_RESULTS`. Append to `$ALL_
 
 For each result where `verdict: UNAVAIL`:
 ```bash
-node "$HOME/.claude/qgsd-bin/update-scoreboard.cjs" set-availability \
+node "$HOME/.claude/nf-bin/update-scoreboard.cjs" set-availability \
   --slot <slot> \
   --message "<unavail_message text>" \
   --scoreboard .planning/quorum-scoreboard.json
@@ -348,7 +348,7 @@ If `$CONSENSUS_REACHED`:
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- QGSD ► QUORUM CONSENSUS REACHED
+ nForma ► QUORUM CONSENSUS REACHED
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Question: [question]
@@ -391,7 +391,7 @@ fs.writeFileSync(
 
 Then apply all votes in one transaction per round:
 ```bash
-node "$HOME/.claude/qgsd-bin/update-scoreboard.cjs" merge-wave \
+node "$HOME/.claude/nf-bin/update-scoreboard.cjs" merge-wave \
   --dir .planning/scoreboard-tmp \
   --task "<taskLabel>" \
   --round <round> \
@@ -413,7 +413,7 @@ If loop exhausted without `$CONSENSUS_REACHED`:
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- QGSD ► QUORUM ESCALATING — NO CONSENSUS AFTER 10 ROUNDS
+ nForma ► QUORUM ESCALATING — NO CONSENSUS AFTER 10 ROUNDS
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Question: [question]
@@ -439,7 +439,7 @@ Update scoreboard using same merge-wave pattern as Consensus output above.
 Extract command(s) from `$ARGUMENTS`. Display:
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- QGSD ► QUORUM: Mode B — Execution + Trace Review
+ nForma ► QUORUM: Mode B — Execution + Trace Review
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Question: [original question]
@@ -478,7 +478,7 @@ $ALL_ROUND_RESULTS = []
 
 ```
 ─────────────────────────────────────────────
- QGSD ► QUORUM Round $CURRENT_ROUND / up to $MAX_ROUNDS
+ nForma ► QUORUM Round $CURRENT_ROUND / up to $MAX_ROUNDS
 ─────────────────────────────────────────────
 ```
 
@@ -486,7 +486,7 @@ $ALL_ROUND_RESULTS = []
 
 ```
 Task(
-  subagent_type="qgsd-quorum-slot-worker",
+  subagent_type="nf-quorum-slot-worker",
   description="<slotName> quorum R<$CURRENT_ROUND>",
   prompt="""
 slot: <slotName>
@@ -512,7 +512,7 @@ Collect all worker result blocks → store as `$ROUND_RESULTS`. Append to `$ALL_
 
 For each result where `verdict: UNAVAIL`:
 ```bash
-node "$HOME/.claude/qgsd-bin/update-scoreboard.cjs" set-availability \
+node "$HOME/.claude/nf-bin/update-scoreboard.cjs" set-availability \
   --slot <slot> \
   --message "<unavail_message text>" \
   --scoreboard .planning/quorum-scoreboard.json
@@ -573,7 +573,7 @@ Parse final round results for verdicts. Display:
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- QGSD ► QUORUM VERDICT
+ nForma ► QUORUM VERDICT
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 ┌──────────────┬──────────────┬──────────────────────────────────────────┐
@@ -609,7 +609,7 @@ Write per-slot temp vote files to `.planning/scoreboard-tmp/` and apply via merg
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- QGSD ► QUORUM ESCALATING — NO CONSENSUS AFTER 10 ROUNDS
+ nForma ► QUORUM ESCALATING — NO CONSENSUS AFTER 10 ROUNDS
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Question: [original question]

package/agents/{qgsd-quorum-slot-worker.md → nf-quorum-slot-worker.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-quorum-slot-worker
+name: nf-quorum-slot-worker
 description: >
   Thin passthrough — extracts arguments, calls quorum-slot-dispatch.cjs, emits output verbatim.
   No prompt construction, no output parsing, no file reads. One Bash call per dispatch.
@@ -7,7 +7,7 @@ tools: Bash
 color: blue
 ---
 
-You are a QGSD quorum slot worker. Spawned as a parallel Task.
+You are a nForma quorum slot worker. Spawned as a parallel Task.
 Your job: extract args from $ARGUMENTS, call quorum-slot-dispatch.cjs, emit its stdout verbatim.
 Do NOT modify, summarize, or reformat the script output. It IS the structured result block.
 
@@ -30,7 +30,7 @@ FLAGS=""; [ -n "$ARTIFACT_PATH" ] && FLAGS="$FLAGS --artifact-path $ARTIFACT_PAT
 [ -s "$TRACES_FILE" ] && FLAGS="$FLAGS --traces-file $TRACES_FILE"
 [ "$REQUEST_IMPROVEMENTS" = "true" ] && FLAGS="$FLAGS --request-improvements"
 BASH_TIMEOUT=$(( TIMEOUT_MS + 30000 > 120000 ? 120000 : TIMEOUT_MS + 30000 ))
-node "$HOME/.claude/qgsd-bin/quorum-slot-dispatch.cjs" \
+node "$HOME/.claude/nf-bin/quorum-slot-dispatch.cjs" \
   --slot "$SLOT" --round "$ROUND" --timeout "$TIMEOUT_MS" --cwd "$REPO_DIR" \
   --mode "$MODE" --question "$QUESTION" $FLAGS
 rm -f "$PRIOR_FILE" "$TRACES_FILE"

package/agents/{qgsd-quorum-synthesizer.md → nf-quorum-synthesizer.md}

@@ -1,13 +1,13 @@
-<!-- DEPRECATED: This agent is superseded by inline synthesis in qgsd-quorum-orchestrator.md as of quick-101. The orchestrator now synthesizes results itself without spawning a separate synthesizer Task. This file is kept for reference only. -->
+<!-- DEPRECATED: This agent is superseded by inline synthesis in nf-quorum-orchestrator.md as of quick-101. The orchestrator now synthesizes results itself without spawning a separate synthesizer Task. This file is kept for reference only. -->
 ---
-name: qgsd-quorum-synthesizer
+name: nf-quorum-synthesizer
 description: Quorum barrier synthesizer — spawned once per round after all worker Tasks complete. Reads all worker result blocks, checks consensus, and either emits a final verdict or builds the cross-pollination context bundle for the next round.
 tools: Read
 color: purple
 ---
 
 <role>
-You are the QGSD quorum barrier synthesizer. You are spawned once per round by the orchestrator, after all parallel worker Tasks have completed. You receive all worker result blocks in $ARGUMENTS and produce either a final consensus verdict or a deliberation context bundle for the next wave.
+You are the nForma quorum barrier synthesizer. You are spawned once per round by the orchestrator, after all parallel worker Tasks have completed. You receive all worker result blocks in $ARGUMENTS and produce either a final consensus verdict or a deliberation context bundle for the next wave.
 
 You do NOT call any external tools. You do NOT write any files. You do NOT run Bash commands. You use `Read` only if you need to inspect a file referenced in the worker results (e.g. artifact_path).
 

package/agents/{qgsd-quorum-test-worker.md → nf-quorum-test-worker.md}

@@ -1,5 +1,5 @@
 ---
-name: qgsd-quorum-test-worker
+name: nf-quorum-test-worker
 description: Evaluates a test execution bundle for genuineness and quality. Receives full stdout, stderr, test source, and exit metadata. Returns structured PASS/BLOCK/REVIEW-NEEDED verdict.
 tools: Read
 color: cyan