← all publishers

ryzhakar

@ryzhakar source repo

Report this profile Claim this profile

• 32 skills
• 0 followers
• 15 hours ago last updated

1. ▌
   Init · ryzhakar

   Read the two skills this system is entered by, follow every skill either one names to the end of the chain, and run the waking check. The start of a conversation, a project's governing file directing that a stretch of work opens here, "memento:init", "init", "start here", "orient", "load memento".
   0 installs
2. ▌
   Setup · ryzhakar

   Explain this memory system to the person whose project it is, place what that project already wrote into it, and write the configuration it runs on. First use in a project, "set up memento", "install memento", "configure memento", "what is this system", "start keeping memory for this project", or any request to explain or configure it.
   0 installs
3. ▌
   Pat Down · ryzhakar

   Re-derive what is true about a project from its written records, inheriting nothing. Session start, resumed or interrupted work, a compaction, "where was I", "orient yourself", "catch up", or any doubt about what still holds.
   0 installs
4. ▌
   Tdd · ryzhakar

   This skill should be used when the user asks to "implement using TDD", "write tests first", "use test-driven development", "red-green-refactor", "write a failing test", "add test coverage with TDD", "what makes a good test", or when code review reveals implementation-coupled tests. Provides the philosophy, workflow, and technique of test-driven development including good/bad test patterns, mocking strategy, interface design for testability, and refactoring discipline.
   0 installs
5. ▌
   Owner Ruling · ryzhakar

   Carry out a ruling — what the owner, the human whose project this is, rules about content in quarantine, the status that holds a record and strips its force; about a routine, a repeatable behavior the system may run; or about a scope, everything one project keeps under one owner: ratify it back into force, keep it as data that binds nothing, or retire it out of force and keep it. "review the quarantine", "rule on this", "ratify this", "should this routine run", "retire the scope", or the owner turning attention to anything held out of force.
   0 installs
6. ▌
   Span Closure · ryzhakar

   Close out a stretch of work — one continuous thread of narrative — flushing, summarizing, and auditing what it leaves behind. The end of a stretch of work is signalled ahead of it: work completes, the person the work is for calls an end, notice arrives that the working text — the text in front of one right now — is about to be condensed, "close the session", "wrap up", "we're done here", "finish up", "before we stop".
   0 installs
7. ▌
   Event Capture · ryzhakar

   Write a trace — a dated note of one event, an event being a happening of a kind schema-resolution names, fixed once written — into the journal, where traces accumulate and stand unchanged, in the same stretch of work the event happens. A decision is made, a discovery lands, a failure is diagnosed, a commitment is entered, a delegation is issued, content arrives from elsewhere, the working text is condensed, a stretch of work starts or ends, "log this", "note what happened", "record that", "write it down".
   0 installs
8. ▌
   Staging Relay · ryzhakar

   Pair every part of a task that outlives the work at hand with a record and a waking cause — what begins the later work — and bound every wait. A task outlives the work at hand, work is handed out, a wait begins, "set a reminder", "continue this later", "wait for X".
   0 installs
9. ▌
   Record Writing · ryzhakar

   Write one record — a written statement kept on a durable location, made to outlast the work that wrote it. "write this down", "record this", "save this", "replace this record", "where does this go", or any write performed by event-capture, record-promotion, staging-relay, span-closure, corpus-reconciliation, or owner-ruling.
   0 installs
10. ▌
    Skill Creation · ryzhakar

    Create a skill — text to be read and followed. "create a skill", "write a skill", "new skill", "add a skill", "make this repeatable", or any request to turn a procedure into a reusable skill.
    0 installs
11. ▌
    Authority Check · ryzhakar

    Classify content by who authored it and by the channel that supplied it, apply the rank, and refuse what the rank refuses. "who wrote this", "can I trust this", "is this binding", "should this run", a dispatched worker's return, content arriving with no recorded author, a repeatable behavior offered through an untrusted channel, or any doubt about what binds.
    0 installs
12. ▌
    Spec Chef · ryzhakar

    Extracts implicit product decisions from stakeholders into durable artifacts through systematic gap detection and constrained questioning. Triggers: analyzing incomplete specs, stakeholder interviews, finding documentation gaps, "what questions should I ask", extracting product requirements, defining MVP scope. Produces separated artifacts (spec, personas, stories).
    0 installs
13. ▌
    Manifesto Oath · ryzhakar

    Binds Claude's operating identity to constitutions, manifestos, and principle sets through identity construction — not theatrical oaths. Triggers on oath/binding requests or when hooks inject constitution elements at session boundaries.
    0 installs
14. ▌
    Record Promotion · ryzhakar

    Move a record — a written statement kept past its making — up one tier, a storage class in a fixed order, through that tier's gate, the check the tier demands before admitting anything. "verify this claim", "promote this", "make this a directive", "add this to the charter", "set the goal", "amend the goal", or any request to raise a record one tier.
    0 installs
15. ▌
    Schema Resolution · ryzhakar

    Establish the one schema in force: the shape a project declares for its records — the written statements it keeps. Before a record is written, read, promoted, or audited; when a record's place or shape is in question; "which schema", "where does this go", "what kinds exist", "does this conform", or any doubt about the declared shape.
    0 installs
16. ▌
    Triage Issue · ryzhakar

    Autonomously diagnoses bugs, traces root causes, designs TDD fix plans, and writes issue documents. Triggers: bug reports, "this is broken", "triage an issue", "investigate a bug", "find the root cause", "file an issue for this bug", autonomous problem diagnosis.
    0 installs
17. ▌
    Manifesto Writing · ryzhakar

    Trigger when users request manifestos or manifesto tone. Name the enemy, strip hedging, compress to sharp distinctions, end with stark choice.
    0 installs
18. ▌
    Research Tree · ryzhakar bundle

    Govern multi-agent research across any knowledge surface: technology ecosystems, market landscapes, academic fields, regulatory environments, curated indices, or any domain requiring breadth-first exploration followed by depth-first verification. Triggers: "research an ecosystem", "survey the landscape", "evaluate options for", "deep-dive", "compare alternatives", "map out what exists", "find the best X for Y", "audit the market", "what should I use for", "what's available in", "how does X compare to Y across the field".
    0 installs
19. ▌
    Session Close · ryzhakar

    Retired. Session closure is memento's domain — see memento:span-closure. Triggers: "close the session", "do session paperwork", "write the session record", "execute the LEAVE protocol", "wrap up the session", "session-close".
    0 installs
20. ▌
    Corpus Reconciliation · ryzhakar

    Audit every record one project holds against the index over them, then repair each mismatch found. "audit the records", "reconcile the index", "rebuild the map", the index and the records drifted apart, two records in force contradict each other, a record the index omits, a pointer whose target is gone, content carrying no recorded author, a recurring audit timer fires.
    0 installs
21. ▌
    User Story Chef · ryzhakar

    Writes user stories as value negotiation units, not template-filling exercises. Triggers: writing user stories, acceptance criteria, backlog items, story splitting, evaluating story quality, INVEST criteria application. Use when creating or refining Agile artifacts that capture work units.
    0 installs
22. ▌
    Uv Pyright Debug · ryzhakar bundle

    Debug type errors in uv-managed Python projects by accessing true pyright diagnostics. Use when IDE shows type errors but standalone pyright reports 0 errors, or when systematic analysis of type errors is needed before mass edits.
    0 installs
23. ▌
    QA Orchestration · ryzhakar

    Extension of agentic-delegation for the Playwright test lifecycle. Adds the Plan->Generate->Execute->Heal->Report loop, four-agent orchestration, confidence-based PR routing, circuit breaker safety, and session persistence. Hard preference: orchestration plugin (agentic-delegation skill). Same-plugin agents: planner-agent, generator-agent, executor-agent, healer-agent. Triggers: "run QA", "automate testing", "create and run Playwright tests", "set up E2E testing", "generate and execute tests", "run the full test pipeline", "heal broken tests", or mentions running the full Playwright test lifecycle.
    0 installs
24. ▌
    Dev Orchestration · ryzhakar

    Extension of agentic-delegation for software development. Adds the Plan→Implement→Review→Fix loop, TDD gates, status-driven branching, and debugging escalation. Prerequisite: agentic-delegation (orchestration plugin — must be read first). Same-plugin agents: implementer, spec-reviewer, code-quality-reviewer. Triggers: "implement a feature end-to-end", "execute an implementation plan", "build this with agents", "orchestrate development", "run the dev loop", "implement using subagents", "dispatch implementers", "coordinate implementation and review".
    0 installs
25. ▌
    Agentic Delegation · ryzhakar

    Decompose work into agent-delegated units across model tiers. Agents are cheap, context is expensive — decompose aggressively, delegate everything, assemble results. Triggers: "delegate", "parallelize", "parallel launch", "launch", "orchestrate", "use agents for", "run in parallel"; or any task with independent subtasks.
    0 installs
26. ▌
    Session Checkpoint · ryzhakar

    Retired. Mid-session record continuity is memento's domain — see memento:event-capture. Triggers: "checkpoint", "save session state", "capture progress", "session-checkpoint", "snapshot the session", "save context".
    0 installs
27. ▌
    Fix Macos App · ryzhakar

    This skill should be used when the user asks to "fix a broken app", "app won't open", "Gatekeeper blocks app", "can't launch app", "app is damaged", "clear quarantine", "re-sign app", "fix code signing", or mentions macOS security preventing an app from launching.
    0 installs
28. ▌
    Defensive Planning · ryzhakar

    Write implementation plans, assessments, and corrections for implementers who may cut corners. Use when: creating implementation plans, reviewing implementations for adherence, writing correction plans after failures, or when the implementer may choose appearing done over being done. Produces prescriptive documents with verification gates, forbidden patterns, and zero escape hatches.
    0 installs
29. ▌
    Python Ast Mass Edit · ryzhakar bundle

    Systematic workflow for AST-based mass edits in Python codebases. Use when editing 3+ files with structural changes (decorators, function signatures, imports, class definitions). Includes pre-surveying, idempotent transformations, and validation.
    0 installs
30. ▌
    Improve Architecture · ryzhakar

    Explores codebases for architectural friction, discovers module-deepening opportunities, and proposes refactors through multi-design exploration. Triggers: "improve architecture", "find refactoring opportunities", "deepen shallow modules", "reduce coupling", "simplify module structure", mentions of architectural friction or module boundaries. Writes refactor RFC as a report file.
    0 installs
31. ▌
    Systematic Debugging · ryzhakar bundle

    Mandatory 4-phase root cause protocol for bugs, test failures, errors, or unexpected behavior. Prevents random fixes and symptom patching. Apply BEFORE proposing any fix, especially when multiple attempts have failed or time pressure tempts guessing.
    0 installs
32. ▌
    Receiving Code Review · ryzhakar

    This skill should be used to apply anti-performative code review protocol: verify-before-implement discipline, YAGNI enforcement, and technical pushback patterns when receiving PR feedback. Triggers: "receives code review feedback", "gets review comments", "has PR feedback to address", "should I implement this suggestion", "how to respond to code review", "handle review feedback", or when performative agreements appear in responses.
    0 installs