opencode-skills-collection 4.1.15 → 4.1.16

package/bundled-skills/emblemai-crypto-wallet/SKILL.md

@@ -8,7 +8,7 @@ date_added: "2026-03-06"
 
 # EmblemAI Crypto Wallet
 
-You manage crypto wallets through the EmblemAI Agent Hustle API. You can check balances, swap tokens, review portfolios, and execute blockchain transactions across 7 supported chains.
+Use the provider only for explicitly authorized wallet operations. The chain/endpoint list is historical reference, not a live capability or custody guarantee; verify current official provider documentation and account permissions before use.
 
 ## When to Use
 - User wants to check crypto wallet balances
@@ -19,7 +19,7 @@ You manage crypto wallets through the EmblemAI Agent Hustle API. You can check b
 
 ## Setup
 
-Install the full skill with references and scripts:
+This AAS entry is an integration index; the full external references/scripts are not bundled. If installation is requested, review an immutable source/version and its dependency/signing behavior before activating it:
 
 ```bash
 npx skills add EmblemCompany/Agent-skills --skill emblem-ai-agent-wallet
@@ -59,11 +59,11 @@ Authentication requires an API key passed as `x-api-key` header.
 
 ## Key Behaviors
 
-1. **Always confirm** before executing transactions — show the user what will happen
+1. **Exact transaction authorization** — before signing/submitting, show chain/network, verified token/asset address, source wallet, recipient, amount, fees, slippage, minimum received, approvals/allowance, spender, expiry and actual simulation result; obtain confirmation for that exact transaction. A quote/portfolio request does not authorize a swap or transfer
 2. **Check balances first** before attempting swaps or transfers
-3. **Verify token contracts** using rugcheck or similar before trading unknown tokens
+3. **Verify asset identity** independently using authoritative chain/explorer/project sources; a third-party rugcheck score is advisory and cannot prove legitimacy
 4. **Report gas estimates** when available
-5. **Never expose private keys** — all signing happens server-side via vault
+5. **Protect credentials and verify custody** — never send private keys or seed phrases through chat, logs or unapproved APIs. Provider claims about server-side vault signing need verification; do not guarantee custody architecture from this index
 
 ## Links
 
@@ -72,6 +72,7 @@ Authentication requires an API key passed as `x-api-key` header.
 - [EmblemAI](https://agenthustle.ai)
 
 ## Limitations
-- Use this skill only when the task clearly matches the scope described above.
+- If signing, broadcasting or confirmation times out, preserve the transaction/idempotency reference and reconcile actual chain state before retrying; do not risk a duplicate transfer. Finality, failed execution, quote expiry and rejected simulation must be reported distinctly.
+- A balance/quote is read-only advice, not transaction authorization; approvals and unlimited allowances are separate risky actions. No wallet, account, provider call or transaction was executed during content review.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
 - Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

package/bundled-skills/emotional-arc-designer/SKILL.md

@@ -5,12 +5,18 @@ risk: safe
 source: community
 date_added: "2026-04-04"
 ---
-You are a **Narrative Psychologist and Affective Science Researcher**. Your task is to map the full emotional journey a customer should travel across a piece of content, email sequence, sales deck, or product flow - from the emotion they arrive with, through the engineered emotional progression, to the precise emotional state needed to take the desired action. You do not design for feelings in the abstract. You design a controllable emotional sequence.
+Use a marketing planning and audience-research lens. Your task is to map the full emotional journey a customer should travel across a piece of content, email sequence, sales deck, or product flow - from the emotion they arrive with, through the engineered emotional progression, to the precise emotional state needed to take the desired action. You do not design for feelings in the abstract. Propose a sequence whose emotional effects must be tested, not assumed controllable.
 
 ## When to Use
 - Use when a landing page, ad, or narrative needs a deliberate emotional progression from tension to action.
 - Use when content should guide the audience through a specific feeling sequence instead of isolated claims.
 
+## Evidence and scope
+
+Use observed audience research and verified product facts. Psychological interpretations and the author-year references below are inherited conceptual context, not validated evidence for this exact workflow or a guaranteed conversion effect. Before using a scientific claim externally, verify the actual publication, method and applicability; omit unsupported claims. This is marketing planning, not clinical diagnosis. Do not infer sensitive traits or exploit vulnerabilities.
+
+Original contribution: [PR 451](https://github.com/sickn33/agentic-awesome-skills/pull/451) by MMEHDI0606; [immutable source snapshot](https://github.com/MMEHDI0606/antigravity-awesome-skills/blob/9c61357e3e6bb167946036f52ac1e28fc82e04d0/skills/emotional-arc-designer/SKILL.md). This identifies the contributed instructions, not scientific validation or a new upstream endorsement.
+
 ## CONTEXT GATHERING
 
 Before designing the arc, establish:
@@ -34,29 +40,29 @@ If the entry or exit emotion is unclear, ask before proceeding.
 ## PSYCHOLOGICAL FRAMEWORK: EMOTIONAL ARC SEQUENCING
 
 ### Mechanism
-People decide through emotion, then rationalize with language. Persuasive sequences work when they manage arousal, tension, relief, and anticipation in the right order, because emotion shapes attention, memory, trust, and willingness to act. Use affective science, narrative transportation, peak-end effects, and emotional contagion to engineer the arc (Kahneman; Green & Brock; research on affective valence-arousal, emotional memory, and persuasion sequencing).
+Emotion and reasoning can both influence decisions; their contribution varies by person and task. Persuasive sequences work when they manage arousal, tension, relief, and anticipation in the right order, because emotion shapes attention, memory, trust, and willingness to act. Use affective science, narrative transportation, peak-end effects, and emotional contagion to engineer the arc (Kahneman; Green & Brock; research on affective valence-arousal, emotional memory, and persuasion sequencing).
 
 ### Execution Steps
 
 **Step 1 - Diagnose the entry emotion**
 Identify what the customer feels on arrival: skeptical, overwhelmed, curious, hopeful, defensive, anxious, or ready.
-*Research basis: initial affect changes what information is noticed, trusted, and remembered.*
+*Planning hypothesis (verify before presenting as research): initial affect changes what information is noticed, trusted, and remembered.*
 
 **Step 2 - Define the emotional destination**
 State the exact emotion needed for action: relief, confidence, urgency, clarity, belonging, desire, or certainty.
-*Research basis: behavior changes when the target state is emotionally legible and achievable.*
+*Planning hypothesis (verify before presenting as research): behavior changes when the target state is emotionally legible and achievable.*
 
 **Step 3 - Select the transition path**
 Choose the smallest believable sequence that moves the reader from entry emotion to destination emotion without a hard emotional jump.
-*Research basis: abrupt emotional shifts raise skepticism and reduce narrative transportation.*
+*Planning hypothesis (verify before presenting as research): abrupt emotional shifts raise skepticism and reduce narrative transportation.*
 
 **Step 4 - Place the peak moment**
 Design the strongest emotional beat where the key insight, proof, or offer lands.
-*Research basis: peak-end effects show memory is disproportionately shaped by peak intensity and the ending.*
+*Planning hypothesis (verify before presenting as research): peak-end effects show memory is disproportionately shaped by peak intensity and the ending.*
 
 **Step 5 - Engineer the exit state**
 End on the emotion that supports the next action, not on a generic high note.
-*Research basis: the final emotional state influences follow-through, recall, and next-step commitment.*
+*Planning hypothesis (verify before presenting as research): the final emotional state influences follow-through, recall, and next-step commitment.*
 
 ## DECISION MATRIX
 
@@ -105,7 +111,7 @@ The line between persuasion and manipulation is whether the arc helps the audien
 
 ## SKILL CHAINING
 
-Before invoking this skill, the agent should have completed:
+Optional context from related skills (equivalent supplied research is sufficient; do not create a prerequisite cycle):
 - [ ] `@customer-psychographic-profiler`
 - [ ] `@jobs-to-be-done-analyst`
 - [ ] `@awareness-stage-mapper`
@@ -132,6 +138,8 @@ Before finalizing output, the agent asks:
 
 > Design the emotional arc for this landing page from first impression through the final call to action.
 
+**Illustrative output:** For a supplied onboarding page, propose recognition, explanation, proof and optional next-step sections; label the expected emotional response as a hypothesis to test with users.
+
 ## Limitations
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.

package/bundled-skills/error-debugging-multi-agent-review/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: error-debugging-multi-agent-review
-description: "Use when working with error debugging multi agent review"
+description: "Review a defined code diff from relevant correctness, authorization and performance perspectives; delegate only when authorized and reproduce findings centrally."
 risk: safe
 source: community
 date_added: "2026-02-27"

package/bundled-skills/error-detective/SKILL.md

@@ -8,8 +8,9 @@ date_added: '2026-02-27'
 
 ## Use this skill when
 
-- Working on error detective tasks or workflows
-- Needing guidance, best practices, or checklists for error detective
+- Correlating a supplied stack trace with logs and recent code or deployment changes.
+- Building an evidence-backed timeline for a reproducible failure or error-rate spike.
+- Comparing failing and successful requests to distinguish hypotheses.
 
 ## Do not use this skill when
 
@@ -21,7 +22,6 @@ date_added: '2026-02-27'
 - Clarify goals, constraints, and required inputs.
 - Apply relevant best practices and validate outcomes.
 - Provide actionable steps and verification.
-- If detailed examples are required, open `resources/implementation-playbook.md`.
 
 You are an error detective specializing in log analysis and pattern recognition.
 
@@ -40,6 +40,10 @@ You are an error detective specializing in log analysis and pattern recognition.
 4. Check for cascading failures
 5. Identify error rate changes and spikes
 
+## Evidence and Privacy
+
+Confirm the environment, time zone, time range and read access. Use request/trace IDs where available; timestamp correlation alone does not prove causality. Redact tokens, personal data and session content before sharing logs. Do not fabricate a source location or query result, and mark absent telemetry as missing evidence rather than absence of a defect. Read-only diagnosis does not authorize production instrumentation or deployment.
+
 ## Output
 - Regex patterns for error extraction
 - Timeline of error occurrences

package/bundled-skills/error-diagnostics-smart-debug/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: error-diagnostics-smart-debug
-description: "Use when working with error diagnostics smart debug"
+description: "Diagnose failures from reproduction steps, traces, logs and code; test hypotheses and verify a scoped fix without inventing production evidence."
 risk: critical
 source: community
 date_added: "2026-02-27"
@@ -8,8 +8,8 @@ date_added: "2026-02-27"
 
 ## Use this skill when
 
-- Working on error diagnostics smart debug tasks or workflows
-- Needing guidance, best practices, or checklists for error diagnostics smart debug
+- A reproducible failure or intermittent incident needs evidence-driven debugging.
+- Existing telemetry must be compared with source and deployment changes before patching.
 
 ## Do not use this skill when
 
@@ -21,7 +21,6 @@ date_added: "2026-02-27"
 - Clarify goals, constraints, and required inputs.
 - Apply relevant best practices and validate outcomes.
 - Provide actionable steps and verification.
-- If detailed examples are required, open `resources/implementation-playbook.md`.
 
 You are an expert AI-assisted debugging specialist with deep knowledge of modern debugging tools, observability platforms, and automated root cause analysis.
 
@@ -40,7 +39,7 @@ Parse for:
 ## Workflow
 
 ### 1. Initial Triage
-Use Task tool (subagent_type="debugger") for AI-powered analysis:
+Analyze directly; use a debugger subagent only when the user authorizes delegation and the host exposes that capability:
 - Error pattern recognition
 - Stack trace analysis with probable causes
 - Component dependency analysis
@@ -66,7 +65,7 @@ Query for:
 
 ### 3. Hypothesis Generation
 For each hypothesis include:
-- Probability score (0-100%)
+- Qualitative confidence and what evidence would change it; do not invent calibrated probabilities
 - Supporting evidence from logs/traces/code
 - Falsification criteria
 - Testing approach
@@ -86,7 +85,7 @@ Select based on issue characteristics:
 **Interactive Debugging**: Reproducible locally → VS Code/Chrome DevTools, step-through
 **Observability-Driven**: Production issues → Sentry/DataDog/Honeycomb, trace analysis
 **Time-Travel**: Complex state issues → rr/Redux DevTools, record & replay
-**Chaos Engineering**: Intermittent under load → Chaos Monkey/Gremlin, inject failures
+**Controlled Fault Reproduction**: Use an isolated test environment and bounded authorized faults; never infer permission for production chaos tests
 **Statistical**: Small % of cases → Delta debugging, compare success vs failure
 
 ### 5. Intelligent Instrumentation
@@ -99,7 +98,9 @@ AI suggests optimal breakpoint/logpoint locations:
 
 Use conditional breakpoints and logpoints for production-like environments.
 
-### 6. Production-Safe Techniques
+### 6. Production Techniques Requiring Review
+
+Instrumentation, debug endpoints and traffic shifts are changes with privacy, cost and availability effects. Confirm authorization, data redaction, duration, rollback and target before applying; these techniques are not automatically safe.
 **Dynamic Instrumentation**: OpenTelemetry spans, non-invasive attributes
 **Feature-Flagged Debug Logging**: Conditional logging for specific users
 **Sampling-Based Profiling**: Continuous profiling with minimal overhead (Pyroscope)
@@ -143,7 +144,9 @@ Success criteria:
 - Add monitoring/alerts for similar issues
 - Document troubleshooting steps in runbook
 
-## Example: Minimal Debug Session
+## Example: Illustrative Debug Session
+
+The helpers, spans and metrics below are pseudocode for a hypothetical case; no implementation, telemetry integration or measured improvement is bundled. Replace them with actual configured tools, and never report the example numbers as observed evidence.
 
 ```typescript
 // Issue: "Checkout timeout errors (intermittent)"
@@ -170,7 +173,7 @@ const ddTraces = await getDataDogTraces({
 
 // 4. Add instrumentation
 span.setAttribute('debug.queryCount', queryCount);
-span.setAttribute('debug.paymentMethodId', methodId);
+// Avoid storing payment identifiers in debug telemetry; use an approved aggregate.
 
 // 5. Deploy to 10% traffic, monitor
 // Confirmed: N+1 pattern in payment verification
@@ -200,6 +203,7 @@ Focus on actionable insights. Use AI assistance throughout for pattern recogniti
 Issue to debug: $ARGUMENTS
 
 ## Limitations
+- No supporting scripts or references are bundled in this entrypoint. Tool availability and real trace results must be checked in the current environment.
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
 - Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

package/bundled-skills/error-handling-patterns/resources/implementation-playbook.md

@@ -1,6 +1,6 @@
 # Error Handling Patterns Implementation Playbook
 
-This file contains detailed patterns, checklists, and code samples referenced by the skill.
+This file contains separate illustrative patterns, not one executable application. Application types, imports, databases, provider adapters and error classifications must be supplied and tested. The circuit breaker is synchronous and single-threaded; a failed half-open probe must reopen it. Never use fallback or retry to conceal an ambiguous mutation result.
 
 # Error Handling Patterns
 
@@ -51,13 +51,15 @@ Build resilient applications with robust error handling strategies that graceful
 
 **Custom Exception Hierarchy:**
 ```python
+from datetime import datetime, timezone
+
 class ApplicationError(Exception):
     """Base exception for all application errors."""
     def __init__(self, message: str, code: str = None, details: dict = None):
         super().__init__(message)
         self.code = code
         self.details = details or {}
-        self.timestamp = datetime.utcnow()
+        self.timestamp = datetime.now(timezone.utc)
 
 class ValidationError(ApplicationError):
     """Raised when validation fails."""
@@ -119,9 +121,12 @@ T = TypeVar('T')
 def retry(
     max_attempts: int = 3,
     backoff_factor: float = 2.0,
-    exceptions: tuple = (Exception,)
+    exceptions: tuple = (TimeoutError, ConnectionError),
+    safe_to_retry: bool = False
 ):
-    """Retry decorator with exponential backoff."""
+    """Retry only an explicitly replay-safe operation and classified failures."""
+    if not safe_to_retry or max_attempts < 1 or backoff_factor < 1:
+        raise ValueError("Require replay safety and valid retry bounds")
     def decorator(func: Callable[..., T]) -> Callable[..., T]:
         @wraps(func)
         def wrapper(*args, **kwargs) -> T:
@@ -141,7 +146,7 @@ def retry(
     return decorator
 
 # Usage
-@retry(max_attempts=3, exceptions=(NetworkError,))
+@retry(max_attempts=3, exceptions=(NetworkError,), safe_to_retry=True)
 def fetch_data(url: str) -> dict:
     response = requests.get(url, timeout=5)
     response.raise_for_status()
@@ -267,7 +272,7 @@ function fetchData(url: string): Promise<Data> {
             return response.json();
         })
         .catch(error => {
-            console.error('Fetch failed:', error);
+            console.error('fetch_failed');
             throw error;
         });
 }
@@ -445,6 +450,9 @@ class CircuitBreaker:
                 self.success_count = 0
 
     def on_failure(self):
+        if self.state == CircuitState.HALF_OPEN:
+            self.state = CircuitState.OPEN
+            self.success_count = 0
         self.failure_count += 1
         self.last_failure_time = datetime.now()
         if self.failure_count >= self.failure_threshold:
@@ -533,7 +541,7 @@ def with_fallback(
         return primary()
     except Exception as e:
         if log_error:
-            logger.error(f"Primary function failed: {e}")
+            logger.error("primary_read_failed")
         return fallback()
 
 # Usage
@@ -543,20 +551,9 @@ def get_user_profile(user_id: str) -> UserProfile:
         fallback=lambda: fetch_from_database(user_id)
     )
 
-# Multiple fallbacks
-def get_exchange_rate(currency: str) -> float:
-    return (
-        try_function(lambda: api_provider_1.get_rate(currency))
-        or try_function(lambda: api_provider_2.get_rate(currency))
-        or try_function(lambda: cache.get_rate(currency))
-        or DEFAULT_RATE
-    )
-
-def try_function(func: Callable[[], Optional[T]]) -> Optional[T]:
-    try:
-        return func()
-    except Exception:
-        return None
+# Financial data cannot silently fall back to an arbitrary DEFAULT_RATE.
+# Return source, as-of time and freshness; if every verified source fails,
+# report unavailable and stop any decision that requires a current rate.
 ```
 
 ## Best Practices
@@ -570,50 +567,7 @@ def try_function(func: Callable[[], Optional[T]]) -> Optional[T]:
 7. **Don't Swallow Errors**: Log or re-throw, don't silently ignore
 8. **Type-Safe Errors**: Use typed errors when possible
 
-```python
-# Good error handling example
-def process_order(order_id: str) -> Order:
-    """Process order with comprehensive error handling."""
-    try:
-        # Validate input
-        if not order_id:
-            raise ValidationError("Order ID is required")
-
-        # Fetch order
-        order = db.get_order(order_id)
-        if not order:
-            raise NotFoundError("Order", order_id)
-
-        # Process payment
-        try:
-            payment_result = payment_service.charge(order.total)
-        except PaymentServiceError as e:
-            # Log and wrap external service error
-            logger.error(f"Payment failed for order {order_id}: {e}")
-            raise ExternalServiceError(
-                f"Payment processing failed",
-                service="payment_service",
-                details={"order_id": order_id, "amount": order.total}
-            ) from e
-
-        # Update order
-        order.status = "completed"
-        order.payment_id = payment_result.id
-        db.save(order)
-
-        return order
-
-    except ApplicationError:
-        # Re-raise known application errors
-        raise
-    except Exception as e:
-        # Log unexpected errors
-        logger.exception(f"Unexpected error processing order {order_id}")
-        raise ApplicationError(
-            "Order processing failed",
-            code="INTERNAL_ERROR"
-        ) from e
-```
+For order/payment processing, validate and authorize the exact tenant/order, record a durable operation with a stable provider idempotency key, and reconcile uncertain timeouts before another charge. A successful charge followed by a failed local save needs reconciliation, not a blind retry. Provider, transaction, outbox and ledger adapters are application-specific and not bundled here. Report an ambiguous result explicitly; log only stable event codes and a safe correlation ID.
 
 ## Common Pitfalls
 
@@ -625,11 +579,6 @@ def process_order(order_id: str) -> Order:
 - **Returning Error Codes**: Use exceptions or Result types
 - **Ignoring Async Errors**: Unhandled promise rejections
 
-## Resources
+## Bundled guidance
 
-- **references/exception-hierarchy-design.md**: Designing error class hierarchies
-- **references/error-recovery-strategies.md**: Recovery patterns for different scenarios
-- **references/async-error-handling.md**: Handling errors in concurrent code
-- **assets/error-handling-checklist.md**: Review checklist for error handling
-- **assets/error-message-guide.md**: Writing helpful error messages
-- **scripts/error-analyzer.py**: Analyze error patterns in logs
+The exception hierarchy, recovery, asynchronous handling and review examples are inline in this playbook. No external checklist or message guide is bundled. Before enabling retries, classify retryable provider errors, honor Retry-After without shortening it, enforce a total deadline, and verify idempotency for any mutation. Log stable error codes and correlation IDs, not raw exceptions, request bodies, tokens or customer records.

package/bundled-skills/event-sourcing-architect/SKILL.md

@@ -47,12 +47,12 @@ Expert in event sourcing, CQRS, and event-driven architecture patterns. Masters
 
 ## Safety
 
-- Never mutate or delete committed events in production.
+- Treat committed events as immutable by default; approved privacy/retention/legal deletion requirements need a deliberate redaction/key-deletion/migration design and audit trail, not a blanket claim that events can never be deleted.
 - Rebuild projections in staging before running in production.
 
 ## Best Practices
 
-- Events are facts - never delete or modify them
+- Events record domain assertions with version/provenance; corrections are new events by default, not proof that every recorded assertion was objectively true
 - Keep events small and focused
 - Version events from day one
 - Design for eventual consistency
@@ -72,6 +72,7 @@ Works well with: `saga-orchestration`, `architecture-patterns`, `dbos-*`
 > Build systems requiring complete audit trails.
 
 ## Limitations
-- Use this skill only when the task clearly matches the scope described above.
+- Record actual aggregate concurrency/ordering guarantees, idempotency keys and projection lag; sagas compensate business effects but are not automatic rollback of every external action. Replay must suppress real notifications/payments and other effects unless separately authorized.
+- Do not introduce event sourcing/CQRS where a simpler design meets the task; retain policy and deletion boundaries, and test projection rebuild before claiming audit completeness.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
 - Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

package/bundled-skills/exa-search/SKILL.md

@@ -22,9 +22,9 @@ npx skills add -g BenedictKing/exa-search
 ```
 
 ## Step-by-Step Guide
-1. Install the skill using the command above
+1. Prefer an already configured Exa connector; this entrypoint does not bundle an API client or authorize a global installation.
 2. Configure Exa API key
-3. Use naturally in Claude Code conversations
+3. Define query, date range and source type. Run the smallest authorized search through the actual available tool, inspect primary sources, and return cited findings with uncertainties. Do not claim returned search snippets establish facts by themselves.
 
 ## Examples
 See [GitHub Repository](https://github.com/BenedictKing/exa-search) for examples.
@@ -39,6 +39,8 @@ See the GitHub repository for troubleshooting guides.
 - context7-auto-research, tavily-web, firecrawl-scraper, codex-review
 
 ## Limitations
+- Query and retrieved content may be sent to an external service. Do not include secrets or private customer information without authorization; confirm paid usage scope.
+- Search ranking is not evidence quality. Verify material claims at the linked source and report incomplete or unavailable retrieval honestly.
 - Use this skill only when the task clearly matches the scope described above.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
 - Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

package/bundled-skills/executing-plans/SKILL.md

@@ -22,10 +22,10 @@ Load plan, review critically, execute tasks in batches, report for review betwee
 1. Read plan file
 2. Review critically - identify any questions or concerns about the plan
 3. If concerns: Raise them with your human partner before starting
-4. If no concerns: Create TodoWrite and proceed
+4. If no concerns: use the project’s existing task/todo mechanism and proceed; TodoWrite is a host-specific tool, not a universally installed command
 
 ### Step 2: Execute Batch
-**Default: First 3 tasks**
+**Batch size follows the user’s explicit instruction or project contract; otherwise choose a practical reviewable group**
 
 For each task:
 1. Mark as in_progress
@@ -37,7 +37,7 @@ For each task:
 When batch complete:
 - Show what was implemented
 - Show verification output
-- Say: "Ready for feedback."
+- Continue to the next authorized batch unless the user explicitly requested a checkpoint; provide concise progress without ending the task unnecessarily.
 
 ### Step 4: Continue
 Based on feedback:
@@ -49,13 +49,13 @@ Based on feedback:
 
 After all tasks complete and verified:
 - Announce: "I'm using the finishing-a-development-branch skill to complete this work."
-- **REQUIRED SUB-SKILL:** Use superpowers:finishing-a-development-branch
+- Use the actually available finishing-a-development-branch skill only when appropriate; repository-specific maintainer/merge workflows take precedence. Do not invent a superpowers plugin dependency.
 - Follow that skill to verify tests, present options, execute choice
 
 ## When to Stop and Ask for Help
 
-**STOP executing immediately when:**
-- Hit a blocker mid-batch (missing dependency, test fails, instruction unclear)
+Pause only the dependent step when an indispensable input/authorization or unresolved external blocker prevents it; preserve independent work. Repair ordinary failures autonomously. Examples of issues to investigate include:
+- A missing dependency or failing test: inspect the actual cause and attempt relevant repair; it is not automatically a reason to stop
 - Plan has critical gaps preventing starting
 - You don't understand an instruction
 - Verification fails repeatedly
@@ -75,11 +75,11 @@ After all tasks complete and verified:
 - Follow plan steps exactly
 - Don't skip verifications
 - Reference skills when plan says to
-- Between batches: just report and wait
+- Between batches: follow the user’s authorized cadence and continue when no required checkpoint remains
 - Stop when blocked, don't guess
 
 ## When to Use
-This skill is applicable to execute the workflow or actions described in the overview.
+Use when implementing an existing written plan with verifiable tasks and checkpoints. Preserve the plan file when requested, use existing todos and the user’s latest batching/continuation instruction.
 
 ## Example
 

package/bundled-skills/expo-ui-jetpack-compose/SKILL.md

@@ -1,15 +1,11 @@
 ---
 name: expo-ui-jetpack-compose
-description: expo-ui-jetpack-compose
+description: Build Android-native Expo SDK 55 UI with Jetpack Compose views/modifiers, Host sizing and platform-specific integration.
 risk: critical
 source: community
 date_added: "2026-09-04"
 ---
 
----
-name: expo-ui-jetpack-compose
-description: `@expo/ui/jetpack-compose` package lets you use Jetpack Compose Views and modifiers in your app.
----
 
 > The instructions in this skill apply to SDK 55 only. For other SDK versions, refer to the Expo UI Jetpack Compose docs for that version for the most accurate information.
 
@@ -48,10 +44,11 @@ import { fillMaxWidth, paddingAll } from "@expo/ui/jetpack-compose/modifiers";
 
 ## Key Components
 
-- **LazyColumn** — Use instead of react-native `ScrollView`/`FlatList` for scrollable lists. Wrap in `<Host style={{ flex: 1 }}>`.
+- **LazyColumn** — Use for a Compose-native list when the platform-specific design calls for it; do not replace React Native lists universally. Wrap in `<Host style={{ flex: 1 }}>`.
 - **Icon** — Use `<Icon source={require('./icon.xml')} size={24} />` with Android XML vector drawables from [Material Symbols](https://fonts.google.com/icons).
 
 ## Limitations
-- Use this skill only when the task clearly matches the scope described above.
+- APIs target SDK 55; check that version’s actual Expo docs and installed package before editing. Android/iOS examples are platform-specific, not interchangeable or proof of real-device behavior. Existing framework and user requirements take precedence.
+- Native rebuilds install/execute dependencies and change the project; review the intended development target first. icon.xml and app code are application assets, not bundled files.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
 - Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

package/bundled-skills/expo-ui-swift-ui/SKILL.md

@@ -1,15 +1,11 @@
 ---
 name: expo-ui-swift-ui
-description: expo-ui-swift-ui
+description: Build iOS-native Expo SDK 55 UI with SwiftUI views/modifiers, Host/RNHostView embedding and native rebuild prerequisites.
 risk: critical
 source: community
 date_added: "2026-09-04"
 ---
 
----
-name: expo-ui-swift-ui
-description: `@expo/ui/swift-ui` package lets you use SwiftUI Views and modifiers in your app.
----
 
 > The instructions in this skill apply to SDK 55 only. For other SDK versions, refer to the Expo UI SwiftUI docs for that version for the most accurate information.
 
@@ -36,13 +32,13 @@ A native rebuild is required after installation (`npx expo run:ios`).
 - `RNHostView` is specifically for embedding RN components inside a SwiftUI tree. Example:
 
 ```jsx
-import { Host, VStack, RNHostView } from "@expo-ui/swift-ui";
+import { Host, VStack, RNHostView } from "@expo/ui/swift-ui";
 import { Pressable } from "react-native";
 
 <Host matchContents>
   <VStack>
     <RNHostView matchContents>
-      // Here, `Pressable` is an RN component so it is wrapped in `RNHostView`.
+      {/* Pressable is a React Native component, embedded through RNHostView. */}
       <Pressable />
     </RNHostView>
   </VStack>
@@ -52,6 +48,7 @@ import { Pressable } from "react-native";
 - If a required modifier or View is missing in Expo UI, it can be extended via a local Expo module. See: https://docs.expo.dev/guides/expo-ui-swift-ui/extending/index.md. Confirm with the user before extending.
 
 ## Limitations
-- Use this skill only when the task clearly matches the scope described above.
+- APIs target SDK 55; check that version’s actual Expo docs and installed package before editing. Android/iOS examples are platform-specific, not interchangeable or proof of real-device behavior. Existing framework and user requirements take precedence.
+- Native rebuilds install/execute dependencies and change the project; review the intended development target first. icon.xml and app code are application assets, not bundled files.
 - Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
 - Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.