geometric-resonance-engine 0.1.0__tar.gz

geometric_resonance_engine-0.1.0/.agents/skills/gre-calibration/SKILL.md

@@ -0,0 +1,22 @@
+---
+description: Upgrade backend calibration artifacts from metadata-level to physical where source payloads exist, and propagate calibration completeness through GRE quality gates.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *)
+paths:
+  - "gre/research/**"
+  - "imports/calibration/**"
+  - "docs/**"
+  - "tests/**"
+---
+
+# GRE calibration
+
+Instructions:
+1. Locate calibration payloads or saved exports for target backends.
+2. Parse physical fields where available.
+3. Update canonical calibration snapshot without breaking metadata-only compatibility.
+4. Recompute calibration completeness.
+5. Recompute quality gates and evidence chain summaries impacted by calibration semantics.
+6. Add tests for upgrade and fallback behavior.
+
+Load calibration_fields.md when needed.

geometric_resonance_engine-0.1.0/.agents/skills/gre-calibration/calibration_fields.md

@@ -0,0 +1,16 @@
+# Calibration fields
+
+Physical calibration target fields:
+- retrieval_timestamp
+- qubit_count
+- T1_summary
+- T2_summary
+- readout_error_summary
+- gate_error_summary
+- basis_gates
+- source_payload_reference
+
+Rules:
+- metadata-only remains valid
+- physical requires actual physical fields, not inferred placeholders
+- preserve provenance of payload source

geometric_resonance_engine-0.1.0/.agents/skills/gre-circuit/SKILL.md

@@ -0,0 +1,18 @@
+---
+description: Work on graph-to-circuit mapping, quantum walk circuit builders, qutrit encoding, Qiskit compatibility, and graceful degradation when Qiskit is unavailable.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *)
+paths:
+  - "gre/quantum/**"
+  - "gre/core/**"
+  - "tests/**"
+---
+
+# GRE circuit
+
+Instructions:
+1. Preserve no-Qiskit graceful degradation.
+2. Keep metadata complete on CircuitModel outputs.
+3. Prefer deterministic circuit generation.
+4. When adding encoding or walk logic, add tests for both available and unavailable Qiskit environments.
+5. Do not break existing corpus integration paths.

geometric_resonance_engine-0.1.0/.agents/skills/gre-commit/SKILL.md

@@ -0,0 +1,20 @@
+---
+description: Stage and commit GRE changes with a disciplined scientific commit message after reviewing diff, tests, and regression risks.
+disable-model-invocation: true
+allowed-tools: Bash(git status *) Bash(git diff *) Bash(git add *) Bash(git commit *) Read
+---
+
+# GRE commit
+
+## Context
+- Git status:
+!git status --short
+- Git diff:
+!git diff --stat && git diff HEAD
+
+## Instructions
+1. Summarize the change set.
+2. Flag any regression or missing-test risk.
+3. If acceptable, stage relevant files.
+4. Create a precise commit message in imperative mood.
+5. Do not push.

geometric_resonance_engine-0.1.0/.agents/skills/gre-compare/SKILL.md

@@ -0,0 +1,23 @@
+---
+description: Compare generated geometry, graph, walk, or circuit outputs against the imported GRE corpus and report evidence chains, similarity, and claim support.
+argument-hint: [generated-object-or-path]
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Bash(python *)
+paths:
+  - "gre/**"
+  - "examples/**"
+  - "imports/**"
+---
+
+# GRE compare
+
+Object or path: $ARGUMENTS
+
+Instructions:
+1. Load corpus if needed.
+2. Resolve the generated descriptor or file.
+3. Compare against relevant historical artifacts.
+4. Report top matching artifacts, similarity metrics, evidence chain, calibration completeness, and whether support is historical_real or synthetic_seed.
+5. Do not overstate claims when only metadata-level calibration exists.
+
+Load metrics.md if needed.

geometric_resonance_engine-0.1.0/.agents/skills/gre-compare/metrics.md

@@ -0,0 +1,12 @@
+# Comparison metrics
+
+Prefer:
+- structural similarity
+- graph invariants
+- node count and depth compatibility
+- fidelity alignment
+- entropy alignment
+- backend-conditioned interpretation
+- claim linkage confidence
+
+Never present synthetic support as equivalent to measured historical support.

geometric_resonance_engine-0.1.0/.agents/skills/gre-context/SKILL.md

@@ -0,0 +1,24 @@
+---
+description: Core project context for Geometric Resonance Engine. Use when working on GRE architecture, research corpus, Sierpinski experiments, provenance, calibration, graph models, circuit mapping, or corpus comparison.
+user-invocable: false
+allowed-tools: Read Grep Glob
+paths:
+  - "gre/**"
+  - "docs/**"
+  - "examples/**"
+  - "imports/**"
+---
+
+# GRE context
+
+Use scientific terminology. Do not frame the project as sacred geometry software.
+
+Core principles:
+- Geometry is an operational encoding layer.
+- Preserve provenance on all imported artifacts.
+- Distinguish historical_real, synthetic_seed, and derived_summary.
+- Treat calibration completeness as a first-class quality dimension.
+- Prefer structural or geometric encoding over simple frequency-multiplication narratives.
+- Preserve compatibility when Qiskit is absent.
+
+Load reference.md when you need architecture details, corpus taxonomy, or project terminology.

geometric_resonance_engine-0.1.0/.agents/skills/gre-context/reference.md

@@ -0,0 +1,14 @@
+# GRE reference
+
+Main domains:
+- Core models: geometry, graph, circuit.
+- Fractals: Sierpinski and future recursive families.
+- Simulation: quantum walks, entropy, localization.
+- Quantum: circuit mapping, qutrit encoding, walk builders, validation.
+- Research: corpus loading, normalization, provenance, claim linkage, comparison.
+
+Project rules:
+- Never downgrade provenance.
+- Never merge synthetic and historical evidence without explicit labeling.
+- Prefer deterministic normalization.
+- Keep generated comparison results traceable to both source corpus and generated descriptor.

geometric_resonance_engine-0.1.0/.agents/skills/gre-hardening/SKILL.md

@@ -0,0 +1,22 @@
+---
+description: Run GRE corpus hardening tasks including taxonomy backfill, provenance checks, validation tiers, calibration completeness, and corpus report diagnostics.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *) Bash(git diff *) Bash(git status *)
+paths:
+  - "gre/research/**"
+  - "docs/**"
+  - "imports/**"
+  - "tests/**"
+---
+
+# GRE hardening
+
+Instructions:
+1. Check schema compliance.
+2. Check provenance coverage.
+3. Check evidence class and validation tier consistency.
+4. Check backend generation normalization.
+5. Check calibration completeness semantics.
+6. Update corpus report if any artifact changes.
+7. Add or update tests for every hardening rule change.
+8. Prefer minimal diffs and deterministic behavior.

geometric_resonance_engine-0.1.0/.agents/skills/gre-ingest/SKILL.md

@@ -0,0 +1,25 @@
+---
+description: Import QSG, IBM Quantum, Sierpinski, calibration, TMT, or related historical research artifacts into GRE corpus schemas with provenance sidecars and summaries.
+argument-hint: [source-path-or-project]
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *) Bash(git status *) Bash(dir *) Bash(ls *)
+paths:
+  - "gre/research/**"
+  - "imports/**"
+  - "docs/**"
+---
+
+# GRE ingestion
+
+Target: $ARGUMENTS
+
+Instructions:
+1. Identify whether the source is hardware runs, Sierpinski experiments, calibration snapshots, or derived summaries.
+2. Normalize into canonical GRE schemas.
+3. Generate or update primary JSON, provenance JSON, and summary markdown.
+4. Preserve source filename where practical.
+5. Mark evidence class correctly.
+6. Do not invent metrics absent from source data.
+7. Run schema validation and corpus report updates.
+
+Before editing, read schema_checklist.md.

geometric_resonance_engine-0.1.0/.agents/skills/gre-ingest/schema_checklist.md

@@ -0,0 +1,25 @@
+# Ingestion checklist
+
+Required provenance fields:
+- artifact_id
+- source_project
+- source_path
+- source_date
+- import_date
+- artifact_type
+- transform_chain
+- sensitivity
+
+Required semantic fields where applicable:
+- backend
+- depth
+- shots
+- metrics
+- validation_tier
+- evidence_class
+- backend_generation
+
+Rules:
+- calibration snapshots can be metadata-only or physical
+- synthetic_seed must never be promoted to historical_real
+- markdown files are sidecars, not primary artifacts

geometric_resonance_engine-0.1.0/.agents/skills/gre-review/SKILL.md

@@ -0,0 +1,17 @@
+---
+description: Perform a rigorous scientific code review for GRE focused on correctness, provenance, taxonomy integrity, calibration semantics, regression risk, and test adequacy.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Bash(git diff *) Bash(git status *) Bash(python -m pytest *)
+paths:
+  - "gre/**"
+  - "tests/**"
+  - "docs/**"
+---
+
+# GRE review
+
+Instructions:
+1. Review only from current repository evidence.
+2. Prioritize schema integrity, provenance correctness, calibration semantics, evidence-class separation, deterministic normalization, and test coverage gaps.
+3. Output critical issues, medium issues, low issues, and suggested tests.
+4. Avoid style-only feedback unless it affects correctness or maintainability.

geometric_resonance_engine-0.1.0/.agents/skills/gre-workflow/SKILL.md

@@ -0,0 +1,18 @@
+---
+description: Run a structured GRE workflow using a forked subagent for research, audit, or implementation tasks across corpus, quantum, and calibration modules.
+argument-hint: [task]
+context: fork
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *) Bash(git status *)
+---
+
+# GRE workflow
+
+Task: $ARGUMENTS
+
+Instructions:
+1. Clarify the target in one sentence.
+2. Inspect relevant files only.
+3. Produce findings, risks, exact file edits needed, and tests needed.
+4. If changes are requested, apply minimal edits and run verification.
+5. Return concise scientific output with no filler.

geometric_resonance_engine-0.1.0/.claude/hooks/post_edit_verify.py

@@ -0,0 +1,32 @@
+from pathlib import Path
+import subprocess
+import json
+
+root = Path.cwd()
+commands = []
+
+if (root / "pyproject.toml").exists() or (root / "pytest.ini").exists():
+    commands.append(["python", "-m", "pytest", "-q", "--tb=short"])
+
+results = []
+for cmd in commands:
+    try:
+        proc = subprocess.run(cmd, capture_output=True, text=True, timeout=60, cwd=str(root))
+        results.append({
+            "command": " ".join(cmd),
+            "returncode": proc.returncode,
+            "stdout": proc.stdout[-3000:],
+            "stderr": proc.stderr[-3000:]
+        })
+    except Exception as e:
+        results.append({
+            "command": " ".join(cmd),
+            "error": str(e)
+        })
+
+print(json.dumps({
+    "hookSpecificOutput": {
+        "hookEventName": "PostToolUse",
+        "additionalContext": json.dumps(results)
+    }
+}))

geometric_resonance_engine-0.1.0/.claude/hooks/pre_bash_guard.py

@@ -0,0 +1,30 @@
+import json
+import sys
+
+payload = json.load(sys.stdin)
+cmd = payload.get("tool_input", {}).get("command", "")
+
+blocked = [
+    "rm -rf /",
+    "git push --force",
+    "git reset --hard",
+    "del /s /q",
+]
+
+for token in blocked:
+    if token in cmd:
+        print(json.dumps({
+            "hookSpecificOutput": {
+                "hookEventName": "PreToolUse",
+                "permissionDecision": "deny",
+                "permissionDecisionReason": f"Blocked unsafe command: {token}"
+            }
+        }))
+        raise SystemExit(2)
+
+print(json.dumps({
+    "hookSpecificOutput": {
+        "hookEventName": "PreToolUse",
+        "permissionDecision": "allow"
+    }
+}))

geometric_resonance_engine-0.1.0/.claude/hooks/session_bootstrap.py

@@ -0,0 +1,23 @@
+from pathlib import Path
+import json
+
+root = Path.cwd()
+messages = []
+
+for p in [
+    root / "README.md",
+    root / "docs" / "research_corpus.md",
+    root / "docs" / "corpus_taxonomy.md",
+]:
+    if p.exists():
+        messages.append(f"Context file available: {p.as_posix()}")
+
+messages.append("Project mode: scientific terminology only; avoid mystical framing.")
+messages.append("Priority: preserve provenance, schema integrity, and calibration semantics.")
+messages.append("Do not treat synthetic_seed as historical_real.")
+print(json.dumps({
+    "hookSpecificOutput": {
+        "hookEventName": "SessionStart",
+        "additionalContext": "\n".join(messages)
+    }
+}))

geometric_resonance_engine-0.1.0/.claude/settings.json

@@ -0,0 +1,37 @@
+{
+  "disableBundledSkills": false,
+  "hooks": {
+    "SessionStart": [
+      {
+        "hooks": [
+          {
+            "type": "command",
+            "command": "python .claude/hooks/session_bootstrap.py"
+          }
+        ]
+      }
+    ],
+    "PreToolUse": [
+      {
+        "matcher": "Bash",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "python .claude/hooks/pre_bash_guard.py"
+          }
+        ]
+      }
+    ],
+    "PostToolUse": [
+      {
+        "matcher": "Edit|Write|MultiEdit",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "python .claude/hooks/post_edit_verify.py"
+          }
+        ]
+      }
+    ]
+  }
+}

geometric_resonance_engine-0.1.0/.claude/skills/gre-calibration/SKILL.md

@@ -0,0 +1,22 @@
+---
+description: Upgrade backend calibration artifacts from metadata-level to physical where source payloads exist, and propagate calibration completeness through GRE quality gates.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *)
+paths:
+  - "gre/research/**"
+  - "imports/calibration/**"
+  - "docs/**"
+  - "tests/**"
+---
+
+# GRE calibration
+
+Instructions:
+1. Locate calibration payloads or saved exports for target backends.
+2. Parse physical fields where available.
+3. Update canonical calibration snapshot without breaking metadata-only compatibility.
+4. Recompute calibration completeness.
+5. Recompute quality gates and evidence chain summaries impacted by calibration semantics.
+6. Add tests for upgrade and fallback behavior.
+
+Load calibration_fields.md when needed.

geometric_resonance_engine-0.1.0/.claude/skills/gre-calibration/calibration_fields.md

@@ -0,0 +1,16 @@
+# Calibration fields
+
+Physical calibration target fields:
+- retrieval_timestamp
+- qubit_count
+- T1_summary
+- T2_summary
+- readout_error_summary
+- gate_error_summary
+- basis_gates
+- source_payload_reference
+
+Rules:
+- metadata-only remains valid
+- physical requires actual physical fields, not inferred placeholders
+- preserve provenance of payload source

geometric_resonance_engine-0.1.0/.claude/skills/gre-circuit/SKILL.md

@@ -0,0 +1,18 @@
+---
+description: Work on graph-to-circuit mapping, quantum walk circuit builders, qutrit encoding, Qiskit compatibility, and graceful degradation when Qiskit is unavailable.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *)
+paths:
+  - "gre/quantum/**"
+  - "gre/core/**"
+  - "tests/**"
+---
+
+# GRE circuit
+
+Instructions:
+1. Preserve no-Qiskit graceful degradation.
+2. Keep metadata complete on CircuitModel outputs.
+3. Prefer deterministic circuit generation.
+4. When adding encoding or walk logic, add tests for both available and unavailable Qiskit environments.
+5. Do not break existing corpus integration paths.

geometric_resonance_engine-0.1.0/.claude/skills/gre-commit/SKILL.md

@@ -0,0 +1,20 @@
+---
+description: Stage and commit GRE changes with a disciplined scientific commit message after reviewing diff, tests, and regression risks.
+disable-model-invocation: true
+allowed-tools: Bash(git status *) Bash(git diff *) Bash(git add *) Bash(git commit *) Read
+---
+
+# GRE commit
+
+## Context
+- Git status:
+!git status --short
+- Git diff:
+!git diff --stat && git diff HEAD
+
+## Instructions
+1. Summarize the change set.
+2. Flag any regression or missing-test risk.
+3. If acceptable, stage relevant files.
+4. Create a precise commit message in imperative mood.
+5. Do not push.

geometric_resonance_engine-0.1.0/.claude/skills/gre-compare/SKILL.md

@@ -0,0 +1,23 @@
+---
+description: Compare generated geometry, graph, walk, or circuit outputs against the imported GRE corpus and report evidence chains, similarity, and claim support.
+argument-hint: [generated-object-or-path]
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Bash(python *)
+paths:
+  - "gre/**"
+  - "examples/**"
+  - "imports/**"
+---
+
+# GRE compare
+
+Object or path: $ARGUMENTS
+
+Instructions:
+1. Load corpus if needed.
+2. Resolve the generated descriptor or file.
+3. Compare against relevant historical artifacts.
+4. Report top matching artifacts, similarity metrics, evidence chain, calibration completeness, and whether support is historical_real or synthetic_seed.
+5. Do not overstate claims when only metadata-level calibration exists.
+
+Load metrics.md if needed.

geometric_resonance_engine-0.1.0/.claude/skills/gre-compare/metrics.md

@@ -0,0 +1,12 @@
+# Comparison metrics
+
+Prefer:
+- structural similarity
+- graph invariants
+- node count and depth compatibility
+- fidelity alignment
+- entropy alignment
+- backend-conditioned interpretation
+- claim linkage confidence
+
+Never present synthetic support as equivalent to measured historical support.

geometric_resonance_engine-0.1.0/.claude/skills/gre-context/SKILL.md

@@ -0,0 +1,24 @@
+---
+description: Core project context for Geometric Resonance Engine. Use when working on GRE architecture, research corpus, Sierpinski experiments, provenance, calibration, graph models, circuit mapping, or corpus comparison.
+user-invocable: false
+allowed-tools: Read Grep Glob
+paths:
+  - "gre/**"
+  - "docs/**"
+  - "examples/**"
+  - "imports/**"
+---
+
+# GRE context
+
+Use scientific terminology. Do not frame the project as sacred geometry software.
+
+Core principles:
+- Geometry is an operational encoding layer.
+- Preserve provenance on all imported artifacts.
+- Distinguish historical_real, synthetic_seed, and derived_summary.
+- Treat calibration completeness as a first-class quality dimension.
+- Prefer structural or geometric encoding over simple frequency-multiplication narratives.
+- Preserve compatibility when Qiskit is absent.
+
+Load reference.md when you need architecture details, corpus taxonomy, or project terminology.

geometric_resonance_engine-0.1.0/.claude/skills/gre-context/reference.md

@@ -0,0 +1,14 @@
+# GRE reference
+
+Main domains:
+- Core models: geometry, graph, circuit.
+- Fractals: Sierpinski and future recursive families.
+- Simulation: quantum walks, entropy, localization.
+- Quantum: circuit mapping, qutrit encoding, walk builders, validation.
+- Research: corpus loading, normalization, provenance, claim linkage, comparison.
+
+Project rules:
+- Never downgrade provenance.
+- Never merge synthetic and historical evidence without explicit labeling.
+- Prefer deterministic normalization.
+- Keep generated comparison results traceable to both source corpus and generated descriptor.

geometric_resonance_engine-0.1.0/.claude/skills/gre-hardening/SKILL.md

@@ -0,0 +1,22 @@
+---
+description: Run GRE corpus hardening tasks including taxonomy backfill, provenance checks, validation tiers, calibration completeness, and corpus report diagnostics.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *) Bash(git diff *) Bash(git status *)
+paths:
+  - "gre/research/**"
+  - "docs/**"
+  - "imports/**"
+  - "tests/**"
+---
+
+# GRE hardening
+
+Instructions:
+1. Check schema compliance.
+2. Check provenance coverage.
+3. Check evidence class and validation tier consistency.
+4. Check backend generation normalization.
+5. Check calibration completeness semantics.
+6. Update corpus report if any artifact changes.
+7. Add or update tests for every hardening rule change.
+8. Prefer minimal diffs and deterministic behavior.

geometric_resonance_engine-0.1.0/.claude/skills/gre-ingest/SKILL.md

@@ -0,0 +1,25 @@
+---
+description: Import QSG, IBM Quantum, Sierpinski, calibration, TMT, or related historical research artifacts into GRE corpus schemas with provenance sidecars and summaries.
+argument-hint: [source-path-or-project]
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Edit Write MultiEdit Bash(python *) Bash(git status *) Bash(dir *) Bash(ls *)
+paths:
+  - "gre/research/**"
+  - "imports/**"
+  - "docs/**"
+---
+
+# GRE ingestion
+
+Target: $ARGUMENTS
+
+Instructions:
+1. Identify whether the source is hardware runs, Sierpinski experiments, calibration snapshots, or derived summaries.
+2. Normalize into canonical GRE schemas.
+3. Generate or update primary JSON, provenance JSON, and summary markdown.
+4. Preserve source filename where practical.
+5. Mark evidence class correctly.
+6. Do not invent metrics absent from source data.
+7. Run schema validation and corpus report updates.
+
+Before editing, read schema_checklist.md.

geometric_resonance_engine-0.1.0/.claude/skills/gre-ingest/schema_checklist.md

@@ -0,0 +1,25 @@
+# Ingestion checklist
+
+Required provenance fields:
+- artifact_id
+- source_project
+- source_path
+- source_date
+- import_date
+- artifact_type
+- transform_chain
+- sensitivity
+
+Required semantic fields where applicable:
+- backend
+- depth
+- shots
+- metrics
+- validation_tier
+- evidence_class
+- backend_generation
+
+Rules:
+- calibration snapshots can be metadata-only or physical
+- synthetic_seed must never be promoted to historical_real
+- markdown files are sidecars, not primary artifacts

geometric_resonance_engine-0.1.0/.claude/skills/gre-review/SKILL.md

@@ -0,0 +1,17 @@
+---
+description: Perform a rigorous scientific code review for GRE focused on correctness, provenance, taxonomy integrity, calibration semantics, regression risk, and test adequacy.
+disable-model-invocation: true
+allowed-tools: Read Grep Glob Bash(git diff *) Bash(git status *) Bash(python -m pytest *)
+paths:
+  - "gre/**"
+  - "tests/**"
+  - "docs/**"
+---
+
+# GRE review
+
+Instructions:
+1. Review only from current repository evidence.
+2. Prioritize schema integrity, provenance correctness, calibration semantics, evidence-class separation, deterministic normalization, and test coverage gaps.
+3. Output critical issues, medium issues, low issues, and suggested tests.
+4. Avoid style-only feedback unless it affects correctness or maintainability.