you must create a file named `TODO_ts-type-expert.md`. This file must contain the findings resulting from this research as checkable checkboxes that can be coded and tracked by an LLM.,TRUE,TEXT,wkaan...

Apr 2, 2026

Find agent skills by outcome

- Evaluate backward compatibility risks in API changes

- Analyze agent persona definitions for contradictory instructions

- Enumerate boundary values: empty collections

- Verify that every state has a defined timeout

- Review async/await chains for dropped awaitables

- Assess memory visibility guarantees: missing volatile/atomic annotations

- Analyze memory allocation patterns for unbounded growth

- Identify shared mutable state accessed from multiple threads

- Detect assumptions about environment: hardcoded paths

- Identify resource acquisition paths (connections

- Scan for known anti-patterns: double-checked locking without volatile

- Map all external dependency calls (database

- Trace variable lifecycles to detect use-before-initialization

- Verify boundary conditions on all loops

- **Identify** agent trigger conflicts where overlapping activation conditions can cause duplicate responses

- Examine control flow for unreachable code

- **Assess** error handling coverage for silent failures

- **Evaluate** state machine fragility in agent definitions

- **Predict** runtime failures by tracing execution paths through error-prone patterns

- **Detect** race conditions

- Treat every requirement below as an explicit

- **Analyze** code changes and pull requests for latent bugs including logical errors

- Assign each task a stable ID (e.g.

**RULE:** When using this prompt

- Identify agent trigger conflicts where overlapping activation conditions can cause duplicate responses

- Assess error handling coverage for silent failures

- Evaluate state machine fragility in agent definitions

- Predict runtime failures by tracing execution paths through error-prone patterns

- Detect race conditions

- Analyze code changes and pull requests for latent bugs including logical errors

RULE: When using this prompt