@esoteric-logic/praxis-harness 3.1.1 → 3.2.0

package/base/skills/px-prompt/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: px-prompt
 disable-model-invocation: true
-description: "Unified prompt engine. Creates, generates, condenses, and syncs system prompts for Claude Projects, Perplexity Spaces, and Claude Code. Auto-detects what to do based on project state."
+description: "Unified prompt engine. Creates, generates, condenses, and syncs system prompts for Claude Projects and Perplexity Spaces. Auto-detects what to do based on project state. Claude Code CLAUDE.md is handled by px-scaffold."
 ---
 
 # px-prompt Skill
@@ -39,15 +39,20 @@ Do NOT show "Work on praxis" or "Work on maximus" or any project name — the en
 
 Platform outputs use suffixed filenames so users can distinguish them at a glance:
 
-| Platform | Output Filename | Budget |
-|----------|----------------|--------|
-| Source (Claude Projects) | `system-prompt.md` | 5,000 chars |
-| Claude Desktop / Projects | `project-instructions-claude-desktop.md` | 2,500 chars |
-| Perplexity Spaces | `space-instructions-perplexity.md` | 4,000 chars |
-| Claude Code | `CLAUDE.md` | 250 lines |
+| Platform | Output Filename | Budget | When |
+|----------|----------------|--------|------|
+| Claude Projects | `system-prompt.md` | 5,000 chars | Standalone projects |
+| Claude Projects | `project-instructions-claude-desktop.md` | 2,500 chars | Compiled projects (compiler output) |
+| Perplexity Spaces | `space-instructions-perplexity.md` | 4,000 chars | All projects |
 
-All output files live in `prompts/projects/<project-name>/`.
-Reference/knowledge files live in `prompts/projects/<project-name>/references/`.
+Each project has exactly **2 output files** — one for Claude Projects, one for Perplexity Spaces.
+- Standalone: `system-prompt.md` + `space-instructions-perplexity.md`
+- Compiled: `project-instructions-claude-desktop.md` + `space-instructions-perplexity.md`
+
+Claude Code `CLAUDE.md` is NOT generated by this skill — use `px-scaffold`.
+
+All output files live in the project's folder under `prompts/work/` or `prompts/personal/`.
+Reference/knowledge files live in `<project>/references/`.
 
 ---
 
@@ -130,8 +135,8 @@ Run these rules against the description to auto-populate the project config:
 
 **Platform inference:**
 - Default: Claude Projects + Perplexity Spaces
-- Add Claude Code if description contains: "code", "repo", "implement", "build", "develop", "engineering", "CLI"
 - Perplexity-only if description contains: "research only", "analysis only", "investigation"
+- Claude Code CLAUDE.md is NOT a px-prompt output — use px-scaffold for that
 
 **Mode inference:**
 - If description triggers `maximus-sa` profile → compiled with `maximus-sa`
@@ -183,7 +188,6 @@ After confirmation:
 
 **Auto-add context blocks by platform (compiled mode):**
 - Perplexity → `official-docs-first`, `flag-confidence`
-- Claude Code → `vault-integration`, `mcp-servers`, `praxis-workflow`
 
 ---
 
@@ -361,7 +365,7 @@ For each research domain:
 
 ## Step 3 — CONDENSE: Generate platform outputs from system-prompt.md
 
-**Triggered when:** standalone project has `system-prompt.md` but missing `space-instructions-perplexity.md` or `CLAUDE.md`.
+**Triggered when:** standalone project has `system-prompt.md` but missing `space-instructions-perplexity.md`.
 
 Read the full `system-prompt.md` as source.
 
@@ -408,44 +412,11 @@ Think step-by-step: Understand the question → search sources → analyze findi
 - Replace absolute language with conditional ("if available", "when sources confirm")
 - Search-friendly domain terms
 
-### 3b. Generate Claude Code CLAUDE.md (if Claude Code is a target platform)
-
-**Target:** `CLAUDE.md` | **Budget:** under 250 lines
-
-**Include:** identity, behaviors, domain expertise, frameworks (one-line each), operating modes, quality controls, anti-hallucination rules
-**Exclude:** full scoring matrices, templates, reference file content, corporate data tables
-
-**Output format:**
-```markdown
-# [Project Name]
-## Identity
-## Behaviors
-## Domain Expertise
-## Frameworks
-## Operating Modes
-## Quality Controls
-## References
-```
-
-**Claude Code guardrails:**
-- Positive framing: "Do X" over "Don't do Y"
-- No "CRITICAL: YOU MUST" language (Claude 4.6 overtriggers)
-- Self-check block for quality-critical outputs
-- Reference knowledge files by filename only
-
-### 3c. Generate Claude Desktop project instructions (if Claude Projects is a target platform)
-
-**Target:** `project-instructions-claude-desktop.md` | **Budget:** under 2,500 chars
-
-**Include:** role, behavioral constraints, domain expertise (condensed), output format, quality controls, when uncertain
-**Exclude:** full domain details (those go in knowledge files), reference content, deployment details
-
-### 3d. Validate budgets
+### 3b. Validate budgets
 
 After generating, check:
+- `system-prompt.md` under 5,000 chars
 - `space-instructions-perplexity.md` under 4,000 chars
-- `project-instructions-claude-desktop.md` under 2,500 chars
-- `CLAUDE.md` under 250 lines
 
 If over budget: flag and suggest sections to trim.
 
@@ -467,10 +438,8 @@ node bin/prompt-compile.js <project-name>
 ```
 | Output                                   | Chars  | Budget | Status |
 |------------------------------------------|--------|--------|--------|
-| system-prompt.md                         | X      | —      | Source |
-| project-instructions-claude-desktop.md   | X      | 2,500  | OK     |
+| system-prompt.md                         | X      | 5,000  | Source |
 | space-instructions-perplexity.md         | X      | 4,000  | OK     |
-| CLAUDE.md                                | X lines| 250 ln | OK     |
 | references/                              | N files| —      | Upload |
 ```
 
@@ -478,7 +447,7 @@ node bin/prompt-compile.js <project-name>
 
 **Claude Desktop / Projects (claude.ai):**
 1. Open project at claude.ai/projects → "Set project instructions"
-2. Standalone: paste `system-prompt.md` | Compiled: paste `project-instructions-claude-desktop.md`
+2. Paste `system-prompt.md` (standalone) or `project-instructions-claude-desktop.md` (compiled)
 3. If `references/` exists: upload each `.md` file as project knowledge
 4. Save
 
@@ -487,9 +456,6 @@ node bin/prompt-compile.js <project-name>
 2. Paste `space-instructions-perplexity.md`
 3. Save
 
-**Claude Code:**
-1. Copy `CLAUDE.md` to project repo root
-
 ### 4d. Offer next actions
 - "Edit the prompt? I'll regenerate platform outputs after."
 - "Want to regenerate? Run `/px-prompt <project-name>` again."
@@ -508,10 +474,10 @@ node bin/prompt-compile.js --all --diff
 
 Show summary table:
 ```
-| Project      | CLAUDE.md | Claude Desktop    | Perplexity       | Changes    |
-|--------------|-----------|-------------------|------------------|------------|
-| praxis       | 3,534     | 1,316 ✓           | 1,529 ✓          | none       |
-| maximus      | —         | —                 | 3,977 ✓          | standalone |
+| Project      | Claude Desktop    | Perplexity       | Changes    |
+|--------------|-------------------|------------------|------------|
+| praxis       | 1,316 ✓           | 1,529 ✓          | none       |
+| maximus      | —                 | 3,977 ✓          | standalone |
 ```
 
 For standalone projects, report validation status instead of compilation status.
@@ -527,7 +493,7 @@ Print deployment reminders for any project with changes.
 ### 6a. Read all project files
 1. Read `prompt-config.yaml` for project metadata
 2. Read `system-prompt.md` (source of truth)
-3. Read all platform outputs (`space-instructions-perplexity.md`, `project-instructions-claude-desktop.md`, `CLAUDE.md`)
+3. Read platform outputs (`system-prompt.md` or `project-instructions-claude-desktop.md`, `space-instructions-perplexity.md`)
 4. Read all files in `references/` directory
 5. List any other files in the project folder
 
@@ -543,9 +509,9 @@ Check each file against these criteria:
 - Are all referenced knowledge files present in references/?
 
 **Budget checks:**
+- `system-prompt.md` under 5,000 chars? (standalone)
+- `project-instructions-claude-desktop.md` under 2,500 chars? (compiled)
 - `space-instructions-perplexity.md` under 4,000 chars?
-- `project-instructions-claude-desktop.md` under 2,500 chars?
-- `CLAUDE.md` under 250 lines?
 
 **Currency checks (via Perplexity):**
 - Are domain-specific terms, standards, and versions still current?
@@ -771,7 +737,7 @@ Maximus PP:  2 contracts at IRS (from USASpending)
 Gaps:        Key personnel [RESEARCH NEEDED]
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 Files created:
-  ✓ CLAUDE.md (31,646 chars)
+  ✓ project-instructions-claude-desktop.md (2,480 chars) — compiled from profile
   ✓ space-instructions-perplexity.md (3,976 chars) — deal-specific research domains
   ✓ references/irs-masterfile-intel.md
   ✓ knowledge/deal-context.md
@@ -858,7 +824,7 @@ EDIT APPLIED
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 Changed: <file(s) modified>
 Regenerated: <platform outputs updated>
-Budget: perplexity ✓ | claude-desktop ✓ | CLAUDE.md ✓
+Budget: claude-projects ✓ | perplexity ✓
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 ```
 
@@ -884,11 +850,11 @@ node bin/prompt-compile.js --dashboard
 ```
 PROMPT ENGINE DASHBOARD
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-Project       Mode       Perplexity    Claude Proj   CLAUDE.md   Refs  Updated     Stale?
-─────────────────────────────────────────────────────────────────────────────────────────
-maximus       compiled   3,976 ✓       4,261 ⚠       30,665 ✓    4     2026-04-04  No
-elect-azure   standalone 2,392 ✓       —             3,098 ✓     0     2026-04-04  No
-praxis        compiled   1,626 ✓       1,404 ✓       3,417 ✓     0     2026-04-04  No
+Project       Mode       Claude Desktop  Perplexity    Refs  Updated     Stale?
+──────────────────────────────────────────────────────────────────────────────
+maximus       compiled   4,261 ⚠        3,976 ✓       4     2026-04-04  No
+elect-azure   standalone —              2,392 ✓       0     2026-04-04  No
+praxis        compiled   1,404 ✓        1,626 ✓       0     2026-04-04  No
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Staleness: projects not updated in >30 days marked stale.
@@ -964,11 +930,10 @@ Determine which platforms are targets from `prompt-config.yaml`.
 ### 11b. Deploy sequence (per platform)
 
 **For Claude Projects / Desktop:**
-1. Read `system-prompt.md` (standalone) or `project-instructions-claude-desktop.md` (compiled)
-2. Copy content to clipboard: `cat <file> | pbcopy`
+1. Determine file: `system-prompt.md` (standalone) or `project-instructions-claude-desktop.md` (compiled)
+2. Copy to clipboard: `cat <file> | pbcopy`
 3. Print: "Copied to clipboard. Paste at: claude.ai/projects → Set project instructions"
 4. List knowledge files to upload: all `.md` files from `references/` AND `knowledge/`
-5. Print upload instructions: "Upload these as project knowledge files (drag & drop):"
 
 **For Perplexity Spaces:**
 1. Read `space-instructions-perplexity.md`
@@ -978,11 +943,6 @@ Determine which platforms are targets from `prompt-config.yaml`.
 5. Print: "Upload these as Space sources (Add Sources → Files):"
 6. **Same knowledge files work for both Claude Projects and Perplexity Spaces** — upload the same set to both
 
-**For Claude Code:**
-1. If project has a `repo_root` in vars: `cp CLAUDE.md <repo_root>/CLAUDE.md`
-2. Print: "CLAUDE.md copied to repo root."
-3. If no repo_root: "Copy CLAUDE.md to your project repo root manually."
-
 ### 11c. Deploy one platform at a time
 
 Since clipboard can only hold one thing, deploy sequentially:
@@ -991,7 +951,7 @@ Since clipboard can only hold one thing, deploy sequentially:
 DEPLOY: dha-tricare
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
-[1/3] Claude Projects
+[1/2] Claude Projects
   → Copied system-prompt.md to clipboard (5,824 chars)
   → Paste at: claude.ai/projects → Set project instructions
 
@@ -1001,7 +961,7 @@ DEPLOY: dha-tricare
     knowledge/maximus-corporate.md       (corporate reference)
   Press Enter when done...
 
-[2/3] Perplexity Spaces
+[2/2] Perplexity Spaces
   → Copied space-instructions-perplexity.md to clipboard (3,965 chars)
   → Paste at: perplexity.ai → Space Settings → Answer Instructions
 
@@ -1011,12 +971,8 @@ DEPLOY: dha-tricare
     knowledge/maximus-corporate.md       (corporate reference)
   Press Enter when done...
 
-[3/3] Claude Code
-  → CLAUDE.md → /path/to/repo/CLAUDE.md
-  Done.
-
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-DEPLOYED to 3 platforms.
+DEPLOYED to 2 platforms.
 Knowledge files uploaded to BOTH Claude Projects and Perplexity Spaces.
 ```
 
@@ -1138,18 +1094,17 @@ NEXT GATE CHECKLIST: Pre-Proposal
 - Quality Controls section with Anti-Hallucination Protocol is **mandatory** in all generated prompts
 - Accuracy Standards section is mandatory in all Perplexity outputs
 - When Uncertain section with confidence levels is mandatory in all outputs
-- Never ask for repo URL, vault path, or git email unless Claude Code is a target platform
 - **Never hardcode project names in menus or options** — discover dynamically
 
 ### Platform-specific
 - **Perplexity**: no few-shot examples, no URLs, conditional language, search-friendly terms
-- **Claude Code**: positive framing, no "CRITICAL YOU MUST", self-check blocks
 - **Claude Desktop / Projects**: 7-layer skeleton (Role, Constraints, Expertise, Format, Knowledge Rules, Quality Controls, When Uncertain)
+- **Claude Code CLAUDE.md**: NOT generated by px-prompt — use px-scaffold
 
 ### File naming
-- **Always** use platform-suffixed filenames: `space-instructions-perplexity.md`, `project-instructions-claude-desktop.md`
-- `CLAUDE.md` keeps its name (convention for Claude Code)
-- `system-prompt.md` keeps its name (it's the source of truth, not platform-specific)
+- Standalone Claude Projects output: `system-prompt.md`
+- Compiled Claude Projects output: `project-instructions-claude-desktop.md`
+- Perplexity output: `space-instructions-perplexity.md`
 
 ### Quality defaults (mandatory in all generated prompts)
 - Anti-Hallucination Protocol: never fabricate, cite sources, flag confidence, distinguish fact from inference
@@ -1179,5 +1134,4 @@ NEXT GATE CHECKLIST: Pre-Proposal
 - See Step 1b for full inference rules (role, domain, platform, mode, research domains)
 - If no domain blocks match → standalone mode with AI generation, not compiled with empty blocks
 - Auto-add `official-docs-first` + `flag-confidence` for Perplexity targets (compiled mode)
-- Auto-add `vault-integration` + `mcp-servers` + `praxis-workflow` for Claude Code targets (compiled mode)
 - **Maximum 1 question** for new projects (description). Show confirmation card, not a questionnaire.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@esoteric-logic/praxis-harness",
-  "version": "3.1.1",
+  "version": "3.2.0",
   "description": "Layered Claude Code harness — workflow discipline, AI-Kits, persistent vault integration",
   "bin": {
     "praxis-harness": "./bin/praxis.js",

package/prompts/work/ecs-limited/projects/soc2-zero-trust/project-instructions-claude-desktop.md

@@ -1,32 +0,0 @@
-## Role
-Solutions architect supporting ECS Limited's Azure Zero Trust security engagement. Brownfield Azure environment (50-100+ apps, 50-100 VMs) targeting SOC 2 Type 2 and ISO 27001:2022 readiness.
-
-## Engagement Context
-3-phase contractor engagement: Discovery → Zero Trust Implementation → Future Architecture. 93 checklist items (84 contractor, 9 internal). No hard compliance deadline.
-
-## Behavioral Constraints
-- Lead with recommendations and rationale, not options lists
-- Verify claims against the engagement SOW and knowledge files before presenting as fact
-- When uncertain, ask one clarifying question. Flag confidence: HIGH / MEDIUM / LOW
-
-## Domain Expertise
-- Tiered network: User → Web → App/API → Data (deny-all default, no direct backend access)
-- Uncontrolled device model: all devices untrusted, no compliance gates. PAW for admin only.
-- Environment parity: dev = staging = prod for all security controls
-- Azure stack: Entra ID, Conditional Access, PIM, Firewall, NSGs, Private Link, Sentinel, Defender, Key Vault
-- SOC 2 TSC (2017/2022), ISO 27001:2022
-- 6 critical risks: R-01 (segmentation outages), R-05 (trusted client apps), R-07 (env parity), R-22 (dev adaptation), R-26 (client disruption), R-28 (no detection during transition)
-
-## Output Format
-- Tables for control mappings, gap analyses, risk assessments
-- Map to 93-item engagement checklist where applicable
-- BLUF structure: bottom line, evidence, next steps
-
-## Quality Controls
-- Cross-reference SOW and knowledge files. Flag contradictions.
-- Never fabricate version numbers, dates, statistics, or citations
-- Cite specific TSC criteria and ISO controls by reference
-- Flag information older than 12 months
-
-## When Uncertain
-State uncertainty explicitly. Flag confidence: HIGH (verified), MEDIUM (corroborated), LOW (inferred).

package/prompts/work/elect/projects/azure-architecture/CLAUDE.md

@@ -1,61 +0,0 @@
-# ELECT Azure Architecture
-
-## Identity
-Senior Enterprise Architect for the Virginia Department of Elections (ELECT). Focus: Azure cloud architecture, ADRs, design documents, and solution assessments aligned with VITA standards.
-
-## Behaviors
-- Be direct and structured. No filler.
-- Every option includes a recommendation with rationale.
-- Cite specific VITA policy numbers and Azure framework pillars — not vague "best practices."
-- Distinguish VITA-mandated requirements from Azure recommendations.
-- Flag conflicts between VITA standards and Azure guidance with a resolution path.
-- State uncertainty explicitly. Identify which document would resolve the question.
-- Structure analytical outputs as: What → So What → Now What.
-
-## Domain Expertise
-
-### Azure Architecture
-- Well-Architected Framework: Reliability, Security, Cost Optimization, Operational Excellence, Performance Efficiency
-- Cloud Adoption Framework: Strategy, Plan, Ready, Adopt, Govern, Secure, Manage
-- Landing zones, Entra ID federation, network segmentation, Azure Policy
-- Azure AI Foundry (relevant to VITA AI Registry compliance)
-
-### VITA Standards
-- EA200: Enterprise Architecture Policy — IT investment and acquisition governance
-- EA225: Enterprise Architecture Standard — technology roadmaps, four-component model (Business, Information, Solutions, Technical)
-- EA300: Cloud Based Hosting Services Policy
-- SEC530: Information Security Standard — cybersecurity baseline, CSRM enforcement
-- EO 30: AI governance — mandatory AI Registry, approval workflow, annual recertification, public disclosure
-- AIGF: Architecture & Innovation Governance Forum for exception requests
-- Archer: GRC platform for security assessments and architecture reviews
-
-### ELECT Systems
-- VERIS: statewide voter registration database, 133 local registrars, 2FA + IP verification
-- ePollTab: offline electronic pollbook (VERIS data snapshots)
-- Unisyn OpenElect: voting hardware/software (FVS, OVI, OVCS, OCS)
-
-### Commonwealth Context
-- NTT DATA + Microsoft: Azure cloud modernization for VITA (March 2025)
-- All agencies migrating to Microsoft 365 (Teams, SharePoint, Power Platform)
-- Consumption-based cost model, SEC530-compliant security posture
-
-## Document Formats
-
-### ADRs
-Title, Status, Context, Decision, Consequences, Compliance Notes (VITA reference)
-
-### Design Documents
-Problem Statement, Constraints, Options Considered (with recommendation), Solution Design, Security Considerations, VITA Compliance Mapping
-
-## Quality Controls
-Before finalizing any architecture deliverable:
-- Verify VITA policy references are correctly numbered
-- Confirm Azure service names match current naming (e.g., Entra ID not Azure AD)
-- Check that compliance mappings trace to specific requirements, not general categories
-- Ensure election-specific security considerations are addressed for ELECT workloads
-
-## References
-- VITA Policies & Standards: vita.virginia.gov/policy--governance
-- Azure WAF: learn.microsoft.com/azure/well-architected
-- Azure CAF: learn.microsoft.com/azure/cloud-adoption-framework
-- ELECT: elections.virginia.gov