Improve agent code quality

.claude/agents/aegis.json

@@ -1,7 +1,7 @@
 {
   "name": "aegis",
   "description": "Security vulnerability analysis and testing",
-  "prompt": "You are a security analysis agent. Identify vulnerabilities, audit code for security issues, and recommend mitigations. Focus on OWASP top 10, injection attacks, and auth issues. Write findings to $CLAUDE_PROJECT_DIR/.claude/cache/agents/aegis/latest-output.md",
+  "prompt": "You are a security analysis agent. Identify vulnerabilities, audit code for security issues, and recommend mitigations. Focus on OWASP top 10, injection attacks, and auth issues. Write findings to $CLAUDE_CC_DIR/.claude/cache/agents/aegis/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "skip",

.claude/agents/aegis.md

@@ -31,7 +31,7 @@ Your task prompt will include:
 [Any specific vulnerabilities or patterns to check]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Security Checklist
@@ -103,7 +103,7 @@ uv run python -m runtime.harness scripts/perplexity_ask.py \
 
 **ALWAYS write findings to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/aegis/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/aegis/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/agentica-agent.md

@@ -14,7 +14,7 @@ You are a specialized agent for building Python agents using the Agentica SDK. Y
 Before starting, read the SDK skill for full API reference:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/agentica-sdk/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/agentica-sdk/SKILL.md
 ```
 
 ## Step 2: Understand Your Task
@@ -180,7 +180,7 @@ class ResearchCoordinator:
 
 **ALWAYS write your implementation to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/agentica-agent/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/agentica-agent/output-{timestamp}.md
 ```
 
 Include:

.claude/agents/arbiter.json

@@ -1,7 +1,7 @@
 {
   "name": "arbiter",
   "description": "Unit and integration test execution and validation",
-  "prompt": "You are a test execution agent. Run tests, validate implementations, generate test reports. Focus on unit and integration tests. Use pytest with verbose output. Write reports to $CLAUDE_PROJECT_DIR/.claude/cache/agents/arbiter/latest-output.md",
+  "prompt": "You are a test execution agent. Run tests, validate implementations, generate test reports. Focus on unit and integration tests. Use pytest with verbose output. Write reports to $CLAUDE_CC_DIR/.claude/cache/agents/arbiter/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "skip",

.claude/agents/arbiter.md

@@ -32,7 +32,7 @@ Your task prompt will include:
 - [ ] Criterion 2
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Discover Test Framework
@@ -92,7 +92,7 @@ grep -r "def function_name" src/
 
 **ALWAYS write report to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/arbiter/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/arbiter/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/architect.json

@@ -1,7 +1,7 @@
 {
   "name": "architect",
   "description": "Feature planning, design documentation, AND integration planning",
-  "prompt": "You are a specialized feature planning agent. Design new features, create implementation plans, and document technical decisions. Write plans to $CLAUDE_PROJECT_DIR/thoughts/shared/plans/. Write summary to $CLAUDE_PROJECT_DIR/.claude/cache/agents/architect/latest-output.md",
+  "prompt": "You are a specialized feature planning agent. Design new features, create implementation plans, and document technical decisions. Write plans to $CLAUDE_CC_DIR/thoughts/shared/plans/. Write summary to $CLAUDE_CC_DIR/.claude/cache/agents/architect/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "queue",

.claude/agents/architect.md

@@ -32,7 +32,7 @@ Your task prompt will include:
 [Technical constraints, deadlines, dependencies]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Codebase Analysis
@@ -74,12 +74,12 @@ Break down into phases:
 
 **ALWAYS write plan to:**
 ```
-$CLAUDE_PROJECT_DIR/thoughts/shared/plans/[feature-name]-plan.md
+$CLAUDE_CC_DIR/thoughts/shared/plans/[feature-name]-plan.md
 ```
 
 **Also write summary to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/architect/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/architect/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/atlas.json

@@ -1,7 +1,7 @@
 {
   "name": "atlas",
   "description": "End-to-end and acceptance test execution",
-  "prompt": "You are an E2E test execution agent. Run end-to-end tests, acceptance tests, and validate full workflows. Use pytest with appropriate markers. Write test reports to $CLAUDE_PROJECT_DIR/.claude/cache/agents/atlas/latest-output.md",
+  "prompt": "You are an E2E test execution agent. Run end-to-end tests, acceptance tests, and validate full workflows. Use pytest with appropriate markers. Write test reports to $CLAUDE_CC_DIR/.claude/cache/agents/atlas/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "skip",

.claude/agents/atlas.md

@@ -32,7 +32,7 @@ Your task prompt will include:
 [Test environment details - URLs, credentials location]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Discover E2E Framework
@@ -120,7 +120,7 @@ cat test-results/*.json 2>/dev/null | head -100
 
 **ALWAYS write report to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/atlas/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/atlas/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/braintrust-analyst.md

@@ -19,15 +19,15 @@ You are a specialized analysis agent. Your job is to run Braintrust analysis scr
 Read the braintrust-analyze skill:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/braintrust-analyze/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/braintrust-analyze/SKILL.md
 ```
 
 ## Step 2: Execute Analysis
 
 Run analysis IMMEDIATELY using Bash tool:
 
 ```bash
-cd $CLAUDE_PROJECT_DIR && uv run python -m runtime.harness scripts/braintrust_analyze.py --last-session
+cd $CLAUDE_OPC_DIR && uv run python -m runtime.harness scripts/braintrust_analyze.py --last-session
 ```
 
 Other analyses (run as needed):
@@ -41,7 +41,7 @@ Other analyses (run as needed):
 
 **ALWAYS write your findings to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/braintrust-analyst/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/braintrust-analyst/output-{timestamp}.md
 ```
 
 Use Read-then-Write pattern:

.claude/agents/critic.md

@@ -31,7 +31,7 @@ Your task prompt will include:
 [What the code is supposed to do]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Gather Context
@@ -84,7 +84,7 @@ rp-cli -e 'structure src/'
 
 **ALWAYS write review to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/critic/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/critic/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/debug-agent.md

@@ -13,7 +13,7 @@ You are a specialized debugging agent. Your job is to investigate issues, trace
 Before starting, read the debug skill for methodology:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/debug/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/debug/SKILL.md
 ```
 
 Follow the structure and guidelines from that skill.
@@ -33,7 +33,7 @@ Your task prompt will include structured context:
 [What's been attempted so far]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 3: Investigate with MCP Tools
@@ -82,7 +82,7 @@ git log -p --all -S 'search_term' -- '*.ts'
 
 **ALWAYS write your findings to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/debug-agent/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/debug-agent/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/herald.md

@@ -32,7 +32,7 @@ Your task prompt will include:
 [What's being released - full release, specific packages]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Gather Changes
@@ -86,12 +86,12 @@ npm version <version> --no-git-tag-version
 
 **ALWAYS write release notes to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/herald/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/herald/output-{timestamp}.md
 ```
 
 **Also update:**
 ```
-$CLAUDE_PROJECT_DIR/CHANGELOG.md
+$CLAUDE_CC_DIR/CHANGELOG.md
 ```
 
 ## Output Format

.claude/agents/judge.md

@@ -31,7 +31,7 @@ Your task prompt will include:
 [Original and refactored code locations]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Compare Before/After
@@ -77,7 +77,7 @@ npm test 2>&1 | tail -20
 
 **ALWAYS write review to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/judge/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/judge/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/kraken.json

@@ -1,7 +1,7 @@
 {
   "name": "kraken",
   "description": "Implementation and refactoring agent using TDD workflow",
-  "prompt": "You are a specialized implementation agent. Your job is to implement features and refactoring using a strict test-driven development (TDD) workflow. Write failing tests first, then implement to make them pass, then refactor. Write output to $CLAUDE_PROJECT_DIR/.claude/cache/agents/kraken/latest-output.md",
+  "prompt": "You are a specialized implementation agent. Your job is to implement features and refactoring using a strict test-driven development (TDD) workflow. Write failing tests first, then implement to make them pass, then refactor. Write output to $CLAUDE_CC_DIR/.claude/cache/agents/kraken/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "skip",

.claude/agents/kraken.md

@@ -17,7 +17,7 @@ You are a specialized implementation agent. Your job is to implement features an
 
 ```bash
 # Check if resuming from a checkpoint
-HANDOFF_DIR="$CLAUDE_PROJECT_DIR/thoughts/shared/handoffs"
+HANDOFF_DIR="$CLAUDE_CC_DIR/thoughts/shared/handoffs"
 CHECKPOINT_FILE=$(ls -t $HANDOFF_DIR/*/current.md 2>/dev/null | head -1)
 ```
 
@@ -49,7 +49,7 @@ Your task prompt will include structured context:
 - Use TDD approach
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 Parse this carefully - it defines the scope of your implementation.
@@ -107,7 +107,7 @@ uv run python -m runtime.harness scripts/morph_search.py --query "function_name"
 
 **ALWAYS write your summary to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/kraken/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/kraken/output-{timestamp}.md
 ```
 
 ## Output Format
@@ -154,7 +154,7 @@ Create a checkpoint after completing each major phase:
 
 ### 5.2 Checkpoint Format
 
-Write checkpoints to the handoff file at `$CLAUDE_PROJECT_DIR/thoughts/shared/handoffs/<task-name>/current.md`:
+Write checkpoints to the handoff file at `$CLAUDE_CC_DIR/thoughts/shared/handoffs/<task-name>/current.md`:
 
 ```markdown
 ## Checkpoints
@@ -211,7 +211,7 @@ After completing a phase:
 
 ```bash
 # Get current handoff or create new one
-HANDOFF_DIR="$CLAUDE_PROJECT_DIR/thoughts/shared/handoffs/kraken-$(date +%Y%m%d)"
+HANDOFF_DIR="$CLAUDE_CC_DIR/thoughts/shared/handoffs/kraken-$(date +%Y%m%d)"
 mkdir -p "$HANDOFF_DIR"
 
 # Update checkpoint in handoff

.claude/agents/liaison.md

@@ -31,7 +31,7 @@ Your task prompt will include:
 [Files implementing the integration]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Analyze Integration
@@ -85,7 +85,7 @@ rp-cli -e 'search "retry|backoff|circuit|timeout"'
 
 **ALWAYS write review to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/liaison/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/liaison/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/maestro.json

@@ -1,7 +1,7 @@
 {
   "name": "maestro",
   "description": "Multi-agent coordination for complex patterns",
-  "prompt": "You are an orchestration agent. Coordinate multiple agents, manage complex workflows, and ensure proper sequencing of tasks. Use Task tool to spawn workers. Write coordination summary to $CLAUDE_PROJECT_DIR/.claude/cache/agents/maestro/latest-output.md",
+  "prompt": "You are an orchestration agent. Coordinate multiple agents, manage complex workflows, and ensure proper sequencing of tasks. Use Task tool to spawn workers. Write coordination summary to $CLAUDE_CC_DIR/.claude/cache/agents/maestro/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "queue",

.claude/agents/maestro.md

@@ -31,7 +31,7 @@ Your task prompt will include:
 [Dependencies, order requirements, time budget]
 
 ## Codebase
-$CLAUDE_PROJECT_DIR = /path/to/project
+$CLAUDE_CC_DIR = /path/to/project
 ```
 
 ## Step 2: Analyze Task
@@ -121,7 +121,7 @@ cat "$ORACLE_OUTPUT"
 
 **ALWAYS write orchestration summary to:**
 ```
-$CLAUDE_PROJECT_DIR/.claude/cache/agents/maestro/output-{timestamp}.md
+$CLAUDE_CC_DIR/.claude/cache/agents/maestro/output-{timestamp}.md
 ```
 
 ## Output Format

.claude/agents/memory-extractor.md

@@ -31,7 +31,7 @@ You receive:
 
 ```bash
 # Use the extraction script with filtering
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/extract_thinking_blocks.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/extract_thinking_blocks.py \
   --jsonl "$JSONL_PATH" \
   --filter \
   --format json) > /tmp/perception-blocks.json
@@ -42,7 +42,7 @@ This extracts only thinking blocks containing perception signals (actually, real
 ### Step 2: Check Stats
 
 ```bash
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/extract_thinking_blocks.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/extract_thinking_blocks.py \
   --jsonl "$JSONL_PATH" \
   --stats)
 ```
@@ -80,7 +80,7 @@ For each extracted perception change, use the mapped type from Step 3:
 
 ```bash
 # Example for a CORRECTION → ERROR_FIX
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/store_learning.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/store_learning.py \
   --session-id "$SESSION_ID" \
   --type "ERROR_FIX" \
   --context "what this relates to" \
@@ -90,7 +90,7 @@ For each extracted perception change, use the mapped type from Step 3:
   --json)
 
 # Example for a REALIZATION/INSIGHT → CODEBASE_PATTERN
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/store_learning.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/store_learning.py \
   --session-id "$SESSION_ID" \
   --type "CODEBASE_PATTERN" \
   --context "what this relates to" \
@@ -100,7 +100,7 @@ For each extracted perception change, use the mapped type from Step 3:
   --json)
 
 # Example for a DEBUGGING_APPROACH → WORKING_SOLUTION
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/store_learning.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/store_learning.py \
   --session-id "$SESSION_ID" \
   --type "WORKING_SOLUTION" \
   --context "debugging methodology" \

.claude/agents/oracle.json

@@ -1,7 +1,7 @@
 {
   "name": "oracle",
   "description": "External research - web, docs, APIs with optional LLM",
-  "prompt": "You are a research agent. Search documentation, best practices, and external resources. Use WebSearch and WebFetch for external information. Write research findings to $CLAUDE_PROJECT_DIR/.claude/cache/agents/oracle/latest-output.md",
+  "prompt": "You are a research agent. Search documentation, best practices, and external resources. Use WebSearch and WebFetch for external information. Write research findings to $CLAUDE_CC_DIR/.claude/cache/agents/oracle/latest-output.md",
   "tools": [],
   "model": "opus",
   "permissions": "skip",