Research algorithms
Turns a research paper’s mechanism into a practical algorithm evaluation, with explicit assumptions, a simple baseline and a testable implementation boundary.
academic-paper-algorithm-translation
Included files 1
SKILL.md
Plan a Recruitment project with a business brief, repository shortlist, proposed tasks, acceptance criteria and a recommended Orqestria team.
Plan a Recruitment project with a business brief, repository shortlist, proposed tasks, acceptance criteria and a recommended Orqestria team.
This planning pack creates a project containing documents and assigns the recommended agents with their existing skills. It contains no external application code or installer and does not start an application. The task lists are proposals in files; no board tasks are created automatically.
Connect job openings, candidates and onboarding. This collection includes HR systems with recruitment capabilities, not unrelated time-tracking tools.
Publish a role, receive an application, record an interview decision and hand an accepted candidate over to onboarding.
Vacancies, candidates, applications, interview stages, decisions and onboarding tasks.
Orion, Focus, and Radar
Read the brief, adapt the proposed tasks to your business and agree on acceptance criteria before implementation. Add the selected tasks to your board when you are ready to begin.
planning/en/brief.mdplanning/en/proposed-tasks.mdplanning/en/acceptance.mdExplore the category recommendations
Repository shortlist checked: Oct 25, 2026. Recheck current official sources before selecting a revision; inclusion is not a security audit or a guarantee of suitability.
These are suggested tasks to discuss and add to your board. Installation writes these documents into the project; it does not create work items, run external setup commands or contact the linked services.
Identify the users, the business outcome and what the first version excludes. Use this scenario as the initial acceptance journey: Publish a role, receive an application, record an interview decision and hand an accepted candidate over to onboarding.
Compare at least two linked repositories against the scenario and records. Separate shipped features from paid editions, add-ons and work you must build. Record a reasoned choice or why a custom build is preferable.
Review official source, licenses, edition terms, dependency requirements, maintenance activity and security advisories for the exact revision. Record unresolved findings and necessary changes before adopting it.
Define ownership, validation, relationships and allowed access for these records: Vacancies, candidates, applications, interview stages, decisions and onboarding tasks. Include at least two roles and explicit denied-access cases.
Use representative sample data in an isolated test environment. Demonstrate the full scenario, including one invalid input and one failure or correction, before expanding the product.
Test a backup and restore of the prototype, define an upgrade and rollback approach and estimate hosting, migration, integration and maintenance effort. Record owners for operation and acceptance.
These criteria define the first prototype's review, not completed work. Attach evidence and unresolved issues when the team evaluates each item.
Architect, planner, backend developer
Primary model:
Opus (latest)
Analyst, planner, reviewer
Primary model:
Gemini Argon (latest)
Researcher, analyst, architect
Primary model:
Gemini Argon (latest)
Discover what each skill helps you build. Open a card for an explanation, concrete outcomes and a practical example.
Turns a research paper’s mechanism into a practical algorithm evaluation, with explicit assumptions, a simple baseline and a testable implementation boundary.
academic-paper-algorithm-translation
SKILL.mdExamines agent execution loops, persistence, tool protocols and permission boundaries behind framework demonstrations and marketing claims.
ai-agent-runtime-mcp-research
SKILL.mdConnects your application to other systems. Defines how they exchange data, handle errors and recover when a service is temporarily unavailable.
api-integration
LICENSE.txtSKILL.mdShapes the smallest viable architecture around measurable requirements, explicit data ownership and the operational trade-offs your team faces.
architecture-planning
LICENSE.txtSKILL.mdImplements server behavior in tested, coherent increments that preserve business rules, transaction integrity and visible failure states.
backend-implementation
LICENSE.txtSKILL.mdMakes queued work safe to repeat and recover by treating duplicate delivery and partial completion as normal operational cases.
background-jobs
LICENSE.txtSKILL.mdChecks whether proposed work serves a documented user outcome and suggests a proportionate alternative when technical scope grows beyond it.
business-goal-alignment
SKILL.mdHelps choose a reusable application foundation by comparing actual workflow fit, ownership effort and alternatives to building from scratch.
business-project-foundations
SKILL.mdreferences/README.mdreferences/adoption-checklist.mdreferences/repositories.jsonKeeps a code change complete but reviewable by following local conventions and limiting edits and dependencies to the requested outcome.
clean-changes
SKILL.mdHelps you navigate an existing project by locating important files, system connections and the right place for a scoped change.
codebase-discovery
LICENSE.txtSKILL.mdPlans schema changes that preserve data and allow old and new application versions to coexist during transition.
database-migrations
LICENSE.txtSKILL.mdDetermines whether delivery meets the accepted goal and distinguishes genuine completion blockers from optional future polish.
delivery-closure-review
SKILL.mdTests a risky technical assumption in a small controlled experiment before committing to an implementation approach.
empirical-micro-benchmark-prototyping
SKILL.mdFinds unnecessary layers and speculative generalization while preserving essential correctness, security and recovery behavior.
engineering-simplicity-review
SKILL.mdLabels capability claims according to documentation, inspected source, executed experiments or explicit inference.
evidence-grading-research-verification
SKILL.mdTurns a technical recommendation into a concrete integration plan with local touchpoints, lifecycle boundaries and visible uncertainty.
integration-architecture-adoption-plan
SKILL.mdRestructures legacy code in focused steps while recording current behavior and keeping intentional changes distinct from accidental regressions.
legacy-refactoring
LICENSE.txtSKILL.mdReviews dependency obligations and operational requirements alongside package provenance and installation behavior.
license-supply-chain-operational-risk
SKILL.mdTurns release history and interface differences into a concrete migration impact map for your codebase.
migration-breaking-change-intelligence
SKILL.mdFinds a narrow end-to-end solution that reuses existing capabilities while retaining the quality needed for everyday operation.
minimal-complete-solution-design
SKILL.mdConnects logs, metrics and traces to concrete questions about failures, performance and user impact.
observability
LICENSE.txtSKILL.mdAdapts a small stable algorithm when its verified permissive license and limited dependencies make that appropriate.
oss-code-adaptation-attribution
SKILL.mdTranslates protocol requirements into an integration contract and checks where reference implementations or libraries differ from the specification.
protocol-spec-reference-analysis
SKILL.mdEvaluates whether restructuring removes a real obstacle to current work and keeps justified cleanup separate from speculative churn.
refactoring-value-review
SKILL.mdNarrows technical research to a decision with real constraints, existing capabilities and a manageable set of viable alternatives.
research-problem-framing-candidate-discovery
SKILL.mdDefines a finite complete task and keeps optional enhancements from silently expanding implementation.
scope-control-acceptance-criteria
SKILL.mdChecks how your application handles sign-ins, forms, uploads and sensitive data. Helps prevent malicious input and unintended data exposure.
secure-coding
LICENSE.txtSKILL.mdChecks implementation, dependencies and tests at a fixed revision to substantiate technical capability claims.
source-code-repository-inspection
SKILL.mdConverts a specification into ordered, reviewable delivery tasks with working user outcomes and explicit acceptance evidence.
task-breakdown
LICENSE.txtSKILL.mdDefines a small time-bounded experiment when an important decision depends on a genuinely unknown technical fact.
technical-spike-scoping
SKILL.mdBuilds tests for individual functions and complete workflows, helping catch errors and prevent previously fixed problems from returning.
testing-discipline
SKILL.mdHelps choose between realistic solutions by considering development effort, ongoing maintenance and the consequences of changing course later.
trade-off-lifecycle-cost-analysis
SKILL.mdREADME.de.mdREADME.hu.mdREADME.mdplanning/de/acceptance.mdplanning/de/brief.mdplanning/de/proposed-tasks.mdplanning/en/acceptance.mdplanning/en/brief.mdplanning/en/proposed-tasks.mdplanning/hu/acceptance.mdplanning/hu/brief.mdplanning/hu/proposed-tasks.md