sherpa-skill 2.0.6 → 2.0.7

package/.agents/skills/sherpa/templates/project-state.md

@@ -0,0 +1,84 @@
+# Hackathon Project State
+
+This file is the persistent state for a sherpa-managed project. It lives in `.sherpa/project-state.md` inside the hackathon project's repository.
+
+Any field whose value is not known must be set to `UNKNOWN`. Never invent or guess missing information.
+
+---
+
+## Hackathon Identity
+
+| Field | Value |
+|-------|-------|
+| Hackathon name | UNKNOWN |
+| Hackathon platform (Devpost, MLH, etc.) | UNKNOWN |
+| Hackathon URL | UNKNOWN |
+| Start date/time | UNKNOWN |
+| Deadline date/time | UNKNOWN |
+| Time remaining | UNKNOWN |
+
+## Rules & Judging
+
+| Field | Value |
+|-------|-------|
+| Hackathon rules summary | UNKNOWN |
+| Judging criteria | UNKNOWN |
+| Criteria weights | UNKNOWN |
+| Submission requirements | UNKNOWN |
+
+## Problem & Idea
+
+| Field | Value |
+|-------|-------|
+| Problem statement | UNKNOWN |
+| Target users | UNKNOWN |
+| Selected idea | UNKNOWN |
+| Product concept | UNKNOWN |
+| MVP scope | UNKNOWN |
+| Key differentiators | UNKNOWN |
+
+## Technical
+
+| Field | Value |
+|-------|-------|
+| Tech stack | UNKNOWN |
+| Architecture summary | UNKNOWN |
+| Repository structure | UNKNOWN |
+| AI/ML pipeline (if any) | UNKNOWN |
+| Deployment configuration | UNKNOWN |
+
+## Progress
+
+| Field | Value |
+|-------|-------|
+| Completed work | UNKNOWN |
+| Current work | UNKNOWN |
+| Blocked work | UNKNOWN |
+| Open questions | UNKNOWN |
+| Known weaknesses | UNKNOWN |
+| Risks | UNKNOWN |
+
+## Scoring
+
+| Field | Value |
+|-------|-------|
+| Current estimated score | UNKNOWN |
+| Potential score | UNKNOWN |
+| Largest gaps | UNKNOWN |
+| Highest-priority improvement | UNKNOWN |
+
+## Deliverables
+
+| Field | Value |
+|-------|-------|
+| Demo status | UNKNOWN |
+| PPT status | UNKNOWN |
+| Documentation status | UNKNOWN |
+| Deployment URL | UNKNOWN |
+
+## Metadata
+
+| Field | Value |
+|-------|-------|
+| State last updated | UNKNOWN |
+| Phase | UNKNOWN |

package/.agents/skills/sherpa/templates/red-team-finding.md

@@ -0,0 +1,73 @@
+# Red Team Finding
+
+## Finding ID
+
+## Title
+
+## Attack Surface
+
+PRODUCT | DIFFERENTIATION | TECHNICAL | RELIABILITY | SECURITY | PRIVACY | UX | ACCESSIBILITY | RESPONSIVE | DEMO | PERFORMANCE | EVIDENCE | CREDIBILITY | DEPLOYMENT | OTHER
+
+## Severity
+
+CRITICAL | HIGH | MEDIUM | LOW | INFO | UNKNOWN
+
+## Confidence
+
+HIGH | MEDIUM | LOW | UNKNOWN
+
+## Affected Flow / Component
+
+## Claim Challenged
+
+Claim ID:
+
+Claim text:
+
+Evidence supporting claim:
+
+Evidence type:
+
+Claim confidence:
+
+## Observation
+
+Include evidence status: FACT | INFERENCE | ASSUMPTION | UNKNOWN and observation status: OBSERVED | REPRODUCED | NOT_REPRODUCED | BLOCKED | UNKNOWN.
+
+## Evidence
+
+Reference route/screen, file and line, viewport/device, steps, captured output/artifact, revision, and observed result. Never include unsupported claims.
+
+## Attack / Reproduction Steps
+
+1.
+
+## User Impact
+
+## Judge Impact
+
+## Technical Impact
+
+## Recommended Mitigation
+
+## Required Owner
+
+PHASE 4 — IMPLEMENTATION | PHASE 5 — TESTING / SECURITY | PHASE 7 — UX | PHASE 8 — RED TEAM | HUMAN ARCHITECTURE APPROVAL | FUTURE PHASE | ACCEPTED RISK
+
+## Required Phase
+
+## Architecture Impact
+
+NONE | POSSIBLE | REQUIRED | UNKNOWN
+
+If REQUIRED, include ARCHITECTURE CHANGE REQUIRED, reason, evidence, impact, estimated effort, alternative, risk, and recommendation. Wait for human approval.
+
+## Verification Method
+
+## Status
+
+OPEN | IN_PROGRESS | MITIGATED | VERIFIED | ACCEPTED_RISK | BLOCKED | UNKNOWN | NOT_APPLICABLE
+
+VERIFIED requires actual verification evidence. ACCEPTED_RISK requires explicit human acceptance evidence.
+
+## Verification Evidence

package/.agents/skills/sherpa/templates/red-team-review.md

@@ -0,0 +1,105 @@
+# Red Team Review
+
+## Project
+
+- Repository root:
+- Revision/commit:
+- Selected idea (read-only):
+- Implementation readiness evidence:
+- Phase 2 idea evaluation evidence:
+- Phase 3 project plan evidence:
+- Phase 4 implementation evidence:
+- Phase 5 testing/security evidence:
+- Phase 7 UX evidence:
+- Review date:
+
+---
+
+## Red Team Baseline
+
+- Critical attack surfaces identified:
+- Critical claims register:
+- Time remaining:
+- Time-aware strategy:
+- Review scope:
+- Environment/runtime actually used:
+- Viewports/devices inspected:
+- Accessibility tools/methods:
+- Excluded or blocked scope:
+- Limitations:
+
+For each baseline claim label FACT | INFERENCE | ASSUMPTION | UNKNOWN and cite evidence. Missing details are UNKNOWN.
+
+---
+
+## Critical Claims Register
+
+| Claim ID | Claim | Source | Evidence | Evidence Type | Confidence | Attack Method | Result |
+|----------|-------|--------|----------|---------------|------------|---------------|--------|
+
+Store detailed claims in `.sherpa/red-team/claim-NNN.md`.
+
+---
+
+## Adversarial Scenarios
+
+| Scenario ID | Attack Surface | Perspective | Severity | Confidence | Status |
+|-------------|----------------|-------------|----------|------------|--------|
+
+Store detailed scenarios in `.sherpa/red-team/scenario-NNN.md`.
+
+---
+
+## Red Team Findings
+
+| Finding ID | Title | Category | Severity | Confidence | Flow/Component | Status |
+|------------|-------|----------|----------|------------|----------------|--------|
+
+Store detailed findings in `.sherpa/red-team/finding-NNN.md`.
+
+---
+
+## Mitigation Routing
+
+| Finding ID | Required Owner | Required Phase | Architecture Impact | Status |
+|------------|----------------|----------------|---------------------|--------|
+
+---
+
+## Red Team Quality Gate
+
+NOT_READY | RED_TEAM_IN_PROGRESS | BLOCKED | CONDITIONAL | READY
+
+Current: ___________
+
+Reason against every READY criterion:
+
+- Critical attack surfaces reviewed:
+- No unresolved critical project-breaking finding:
+- Critical demo path challenged:
+- Major claims challenged:
+- High-severity findings resolved or explicitly accepted:
+- Security/UX dependencies clearly routed:
+- Verification evidence for claimed mitigations:
+- Limitations documented:
+- Remaining risks visible:
+
+---
+
+## Recommended Remediation
+
+Include evidence, impact, priority, time estimate/confidence, architecture impact, human decision needed, verification, and regression plan. Mark TEST REQUIRED for work owned by Phase 5.
+
+---
+
+## Recommended Next Action
+
+Exactly one PRIMARY NEXT ACTION:
+
+WHY:
+
+DONE WHEN:
+
+Time remaining:
+
+Blocked by:

package/.agents/skills/sherpa/templates/red-team-scenario.md

@@ -0,0 +1,43 @@
+# Red Team Scenario
+
+## Scenario ID
+
+## Attack Surface
+
+PRODUCT | DIFFERENTIATION | TECHNICAL | RELIABILITY | SECURITY | PRIVACY | UX | ACCESSIBILITY | RESPONSIVE | DEMO | PERFORMANCE | EVIDENCE | CREDIBILITY | DEPLOYMENT | OTHER
+
+## Attacker / Skeptical Judge Perspective
+
+## Preconditions
+
+## Attack
+
+## Expected Failure
+
+## Actual Result
+
+## Evidence
+
+## Severity
+
+CRITICAL | HIGH | MEDIUM | LOW | INFO | UNKNOWN
+
+## Confidence
+
+HIGH | MEDIUM | LOW | UNKNOWN
+
+## Finding
+
+## Mitigation
+
+## Verification Method
+
+## Status
+
+OPEN | IN_PROGRESS | MITIGATED | VERIFIED | ACCEPTED_RISK | BLOCKED | UNKNOWN | NOT_APPLICABLE
+
+## Applicability
+
+APPLICABLE | NOT_APPLICABLE
+
+If NOT_APPLICABLE, include justification.

package/.agents/skills/sherpa/templates/red-team-verification.md

@@ -0,0 +1,31 @@
+# Red Team Verification
+
+## Finding ID
+
+## Original Problem
+
+## Expected Improvement
+
+## Verification Procedure
+
+Include revision, viewport/device, starting conditions, steps, and related focused check.
+
+## Expected Result
+
+## Actual Result
+
+## Status
+
+PASSED | FAILED | BLOCKED | UNKNOWN
+
+## Evidence
+
+Attach or reference direct observations, artifacts, and actual execution output. A planned or proposed fix is not verification.
+
+## Regression Impact
+
+Record affected flows/components and the regression check actually run. If required test work belongs to Phase 5, mark TEST REQUIRED and link its result when available.
+
+## Related Tests
+
+## Recommended Action

package/.agents/skills/sherpa/templates/repo-analysis.md

@@ -0,0 +1,204 @@
+# Repository Analysis
+
+**Analysis ID:** <unique-id>
+**Analysis date:** <UTC date/time>
+**Hackathon:** <name or UNKNOWN>
+
+---
+
+## 1. Analysis Metadata
+
+| Field | Value |
+|-------|-------|
+| Repository path | <path> |
+| Revision / commit | <commit-hash or UNKNOWN> |
+| Analysis scope | TARGETED | EXHAUSTIVE |
+| Analyzed by | <engine or human> |
+| Evidence sources | <list of files inspected, commands run, or human-provided info> |
+
+**Scope declaration:** This analysis is <TARGETED/EXHAUSTIVE>. <explanation of what was and was not inspected>.
+
+---
+
+## 2. Repository Overview
+
+| Field | Value | Evidence Type | Source |
+|-------|-------|---------------|--------|
+| Repository name | | | |
+| Primary language | | | |
+| Project type | | | |
+| Top-level structure | | | |
+
+---
+
+## 3. Project Metadata
+
+| File | Exists | Content summary | Evidence Type |
+|------|--------|-----------------|---------------|
+| package.json | | | |
+| requirements.txt | | | |
+| pyproject.toml | | | |
+| README.md | | | |
+| Other config files | | | |
+
+---
+
+## 4. Dependencies
+
+| Manifest | Dependencies | Install status | Evidence Type | Source |
+|----------|-------------|----------------|---------------|--------|
+| package.json | | UNKNOWN | | |
+| requirements.txt | | UNKNOWN | | |
+| pyproject.toml | | UNKNOWN | | |
+
+**Rule:** Seeing a dependency manifest does NOT mean dependencies are installed. Mark install status as UNKNOWN unless verified.
+
+---
+
+## 5. Top-Level Project Structure
+
+| Path | Type | Purpose | Evidence Type |
+|------|------|---------|---------------|
+| | | | |
+
+---
+
+## 6. Application Entry Points
+
+| Entry point | Path | Exists | Purpose | Evidence Type |
+|-------------|------|--------|---------|---------------|
+| | | | | |
+
+---
+
+## 7. Frontend
+
+| Component | Path | Exists | Status | Evidence Type | Source |
+|-----------|------|--------|--------|---------------|--------|
+| | | | UNKNOWN | | |
+
+---
+
+## 8. Backend
+
+| Component | Path | Exists | Status | Evidence Type | Source |
+|-----------|------|--------|--------|---------------|--------|
+| | | | UNKNOWN | | |
+
+---
+
+## 9. Database / Data Layer
+
+| Component | Path | Exists | Status | Evidence Type | Source |
+|-----------|------|--------|--------|---------------|--------|
+| | | | UNKNOWN | | |
+
+---
+
+## 10. API Layer
+
+| Endpoint | Path | Exists | Operational | Evidence Type | Source |
+|----------|------|--------|-------------|---------------|--------|
+| | | | UNKNOWN | | |
+
+**Rule:** Seeing an API route does NOT mean the API is operational.
+
+---
+
+## 11. AI/ML Layer
+
+| Component | Path | Exists | Status | Evidence Type | Source |
+|-----------|------|--------|--------|---------------|--------|
+| | | | UNKNOWN | | |
+
+---
+
+## 12. Tests
+
+| Test file | Path | Exists | Passing | Evidence Type | Source |
+|-----------|------|--------|---------|---------------|--------|
+| | | | UNKNOWN | | |
+
+**Rule:** Seeing a test file does NOT mean tests pass. Mark passing status as UNKNOWN unless execution evidence exists.
+
+---
+
+## 13. Deployment / Configuration
+
+| Config file | Path | Exists | Deployed | Evidence Type | Source |
+|-------------|------|--------|----------|---------------|--------|
+| | | | UNKNOWN | | |
+
+**Rule:** Seeing deployment configuration does NOT mean deployment works.
+
+---
+
+## 14. Environment Variables
+
+| Variable | Required | Value source | Evidence Type | Source |
+|----------|----------|--------------|---------------|--------|
+| | | | | |
+
+**Rule:** Do not invent environment variable values.
+
+---
+
+## 15. Documentation
+
+| Doc file | Path | Exists | Current | Evidence Type |
+|----------|------|--------|---------|---------------|
+| README.md | | | UNKNOWN | |
+
+---
+
+## 16. Incomplete Areas
+
+| Area | Description | Impact | Evidence Type | Source |
+|------|-------------|--------|---------------|--------|
+| | | | | |
+
+---
+
+## 17. TODOs / Placeholders
+
+| Location | Content | Evidence Type | Source |
+|----------|-----------|---------------|--------|
+| | | | |
+
+---
+
+## 18. Security Observations
+
+| Observation | Severity | Evidence Type | Source |
+|-------------|----------|---------------|--------|
+| | | | |
+
+**Rule:** Do not claim security is complete without evidence.
+
+---
+
+## 19. Deployment Risks
+
+| Risk | Likelihood | Impact | Evidence Type | Source |
+|------|------------|--------|---------------|--------|
+| | | | | |
+
+---
+
+## 20. Unknowns
+
+| Unknown | Why unknown | Impact | Source |
+|---------|-------------|--------|--------|
+| | | | |
+
+---
+
+## 21. Evidence Sources
+
+| Source | Type | What was verified |
+|--------|------|-------------------|
+| | | |
+
+**Valid source types:** repository inspection, actual execution, external verification, explicit human-provided information.
+
+**Invalid source types:** sherpa-generated artifacts, previous AI recommendations without verification.

package/.agents/skills/sherpa/templates/requirement-audit.md

@@ -0,0 +1,15 @@
+# Requirement Audit
+
+**Requirement ID:** <unique ID>  
+**Requirement:** <requirement text>  
+**Source:** <requirement source>  
+**Importance:** <importance level>  
+**Evidence:** <evidence supporting requirement satisfaction>  
+**Evidence Status:** FACT / INFERENCE / ASSUMPTION / UNKNOWN  
+**Status:** PASS / PASS_WITH_LIMITATION / FAIL / BLOCKED / UNKNOWN / NOT_APPLICABLE  
+**Owner:** <owner name or UNKNOWN>  
+**Finding ID:** <related finding ID or UNKNOWN>  
+
+---
+
+**Rule:** UNKNOWN never becomes PASS without new evidence.

package/.agents/skills/sherpa/templates/research-comparison.md

@@ -0,0 +1,124 @@
+# Research Comparison — <research-query-id>
+
+**Evaluation date:** <UTC date/time>  
+**Hackathon:** <name or UNKNOWN>  
+**Deadline:** <date/time or UNKNOWN>  
+**Time remaining:** <hours or UNKNOWN>  
+**Judging criteria:** <list or UNKNOWN>  
+**Criteria weights:** <weights or UNKNOWN>
+
+---
+
+## 1. Candidate Ideas
+
+| # | Idea Name | One-Line Description |
+|---|-----------|---------------------|
+| 1 | | |
+| 2 | | |
+| 3 | | |
+| 4 | | |
+
+---
+
+## 2. Research Evidence Comparison
+
+| Dimension Category | Idea 1 | Idea 2 | Idea 3 | Idea 4 |
+|-------------------|--------|--------|--------|--------|
+| Problem & Solution | | | | |
+| Innovation & Differentiation | | | | |
+| Technical Feasibility | | | | |
+| Implementation Reality | | | | |
+| Demo & Presentation | | | | |
+| Impact & Market | | | | |
+| Competition & Risk Context | | | | |
+| Risk & Evidence Quality | | | | |
+| **Final Score** | | | | |
+| **Feasibility Gate** | PASS/FAIL | PASS/FAIL | PASS/FAIL | PASS/FAIL |
+| **Red-Team Outcome** | KILL/MAJOR/MINOR | KILL/MAJOR/MINOR | KILL/MAJOR/MINOR | KILL/MAJOR/MINOR |
+
+---
+
+## 3. Decision Outcome
+
+**Outcome:** <SELECT | REVISE | REJECT_ALL>  
+**Recommended idea:** <name or "None">
+
+**Why this outcome:**  
+<explanation tied to category scores, feasibility gate, and red-team outcomes>
+
+---
+
+## 4. Recommended Idea Details (if SELECT)
+
+**Strongest advantages:**
+1. <advantage 1 with category and score>
+2. <advantage 2 with category and score>
+3. <advantage 3 with category and score>
+
+**Biggest risks:**
+<red-team findings that survived but remain concerns>
+
+**What must be true for this idea to succeed:**
+<assumptions that must hold>
+
+**Minimum viable version:**
+<scope that can be built in remaining time>
+
+**Highest-value differentiator:**
+<the one thing that makes this win>
+
+**Estimated implementation difficulty:** <low | medium | high>
+
+**Feasibility gate:** PASS | FAIL (composite: X.X/10)
+
+---
+
+## 5. Why Alternatives Were Rejected
+
+| Idea | Reason | Outcome |
+|------|--------|---------|
+| | | |
+| | | |
+
+---
+
+## 6. REVISE Loop (if outcome is REVISE)
+
+**Weaknesses identified:**
+1. <weakness from red-team>
+2. <weakness from red-team>
+
+**Proposed changes:**
+1. <change 1>
+2. <change 2>
+
+**Re-evaluation required:** <yes / no>  
+**Next iteration due:** <date/time or "pending">
+
+---
+
+## 7. Single Primary Next Action
+
+**PRIMARY NEXT ACTION:** <exactly one action>
+
+**Why this is the highest priority:**  
+<explanation based on impact, time, risk, and judging criteria>
+
+**What you need to do:**  
+<specific, actionable steps>
+
+**What completion looks like:**  
+<clear definition of done>
+
+**Time remaining:** <hours or UNKNOWN>  
+**Blocked by:** <blocker or "None">
+
+---
+
+## 8. Human Confirmation
+
+**Human confirmation status:** <pending | confirmed | rejected>  
+**Confirmed by:** <name or "pending">  
+**Confirmed at:** <UTC date/time or "pending">
+
+**Rule:** Only explicit human confirmation can set `selected idea` in `.sherpa/project-state.md`. The engine recommends; the human decides. No automatic overwrite.