danmestas
- 125 skills
- 0 followers
- 12 hours ago last updated
- ▌ Vault Autoresearch · danmestas bundleUse when iteratively researching a topic and filing the synthesis into an Obsidian vault for later ingestion. Output goes to `<vault>/raw/` (NOT directly into wiki pages) so that `vault-ingest` can process it on its own schedule. Triggers: '/vault-autoresearch', 'autoresearch [topic]', 'research [topic] into the vault', 'deep dive into [topic]', 'find everything about [topic] for the vault', 'research and stage'.
- ▌ Evolution Changelog · danmestasUse when the user types "/changelog evolution", "log this evolution", "track applied evolutions", "update EVOLUTION.md", or asks to record what evolution-driven changes have been applied to the repo. Also fires PostToolUse on `git apply` operations against files matching the evolution-report path pattern. Maintains the EVOLUTION.md changelog at the repo root, one date-section per day, one bullet per applied change.
- ▌ Rtk Code Simplifier · danmestasUse when reviewing or refactoring rtk Rust code that feels verbose, over-engineered, or non-idiomatic — applies iterator chains, early returns, Option/Result chaining, and zero-copy patterns while leaving rtk's mandatory constraints (lazy_static regex, .context() chains, fallback arms) intact
- ▌ Rtk Design Patterns · danmestasUse when designing a new rtk filter module or refactoring an existing one — picks between Newtype, Builder, State Machine, Trait Object, RAII, Strategy, and Extension Trait, with rtk-flavored guardrails against premature abstraction
- ▌ Spdd Reasons Canvas · danmestasUse to create or review a Structured-Prompt-Driven Development REASONS Canvas before code generation: Requirements, Entities, Approach, Structure, Operations, Norms, Safeguards.
- ▌ Spy On Bones Session · danmestasAudit how bones behaves on a target project by fingerprinting <project> before and after `bones up`, capturing `bones status` plus any other state bones leaves behind, then watching a live operator's Claude Code session in real time to find patterns where bones is not running properly. Produces a structured findings report (bugs, inconveniences, improvements). Use whenever the user wants to spy on a bones session, audit bones on a project, watch a Claude session running with bones to surface rough edges, run `bones up` on something and report what's broken, or generally "find what bones gets wrong on <X>" — even if they don't say the word "spy" explicitly.
- ▌ Systematic Debugging · danmestas bundleUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes
- ▌ Gh Project Operations · danmestas bundleUse when adding/updating/deleting issues in projects, changing item statuses, bulk operations, archiving items, or managing project boards daily
- ▌ Receiving Code Review · danmestasUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
- ▌ Rtk Security Guardian · danmestasUse when reviewing rtk for command injection, shell escaping, hook injection, malicious output handling, or path traversal — covers the full rtk threat model with detection commands (ripgrep), safe patterns, and incident-response advisory templates
- ▌ Spdd Iterative Review · danmestasUse after SPDD-generated changes to review behavior, prompt/code consistency, architecture boundaries, code quality, and whether to update prompt first or sync prompt after refactor.
- ▌ Finishing A Bones Leaf · danmestasAfter a bones swarm session closes, decide what to do with the open leaf — fan-in to trunk, keep open, or abandon — and (optionally) materialize trunk into the git worktree, push, and open a PR for the swarmed changes. Use when implementation is complete, all tests pass, and you need to integrate the work back into the bones hub trunk.
- ▌ Requesting Code Review · danmestas bundleUse when completing tasks, implementing major features, or before integrating changes to verify work meets requirements
- ▌ Spdd Abstraction First · danmestasUse during SPDD design review before generation to validate entities, responsibilities, interfaces, boundaries, dependencies, and task granularity.
- ▌ Subagent To Subharness · danmestasUse when you would normally dispatch a subagent via the Agent tool but you're operating as the Darkish Factory orchestrator. Translates the muscle memory into subharness dispatch. Maps task shapes to the right harness role, frames the task in caveman-standard, reads worker output back, decides next step.
- ▌ Monitoring The Operator · danmestasLive-monitor an active operator session from a worker pane (typically a stasi/wait-watch worker auditing the parent that spawned it). Use when a worker should watch the operator's transcript JSONL in real time, surface audit-worthy events without polling, distinguish active autonomous runs from hangs, and avoid the chat-Stop feedback loop that wakes the operator on every routine eval. Triggers when the user spawns a wait-watch worker and asks it to "watch what I do", "monitor my session", "audit the orchestrator live", "tell me if I do anything wrong", or any equivalent.
- ▌ Test Driven Development · danmestas bundleUse when implementing any feature or bugfix, before writing implementation code
- ▌ Working With Sesh Tasks · danmestasUse when you are a worker participating in a sesh goal pursuit (SESH_GOAL_ID is set in your environment) and need to find, claim, work on, or complete tasks linked to that goal. Covers the CAS pull protocol, goal-id filtering (pull-discipline so you don't poach tasks tagged to other goals), task state transitions, and how to surface partial progress to the operator. Triggers when you say "let me check the task queue", "pull the next task", "complete this task", "what's left to do", or when you boot into a session with SESH_GOAL_ID exported.
- ▌ Signoz Dashboard Builder · danmestas bundleUse when creating or updating SigNoz dashboards via the MCP API. Triggers on requests to build dashboards, add panels, visualize metrics/logs/traces in SigNoz, or debug dashboard queries that show "No Data" or "Something went wrong". Also use when working with Claude Code telemetry in SigNoz.
- ▌ Dispatching Parallel Agents · danmestasRun N concurrent subagent sessions for parallel debugging or independent work. Discover ready work from the task list; roll up cross-agent status by aggregating results when agents complete. Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies.
- ▌ Subagent Driven Development · danmestas bundleExecute a plan in parallel by dispatching a fresh subagent per task. Each implementer subagent runs in isolated context (no inheritance from the controller). Two-stage review (spec compliance + code quality) per task. Use when the plan has independent tasks that can run concurrently; use executing-plans instead for single-session inline runs.
- ▌ Investigating Agent Sessions · danmestasUse when debugging or auditing another Claude Code session, investigating what a prior agent did or left behind, reconstructing a failure that hit an earlier session, finding the cause of mystery state in a project (orphan processes, dirty files, broken hooks), or building context before continuing someone else's work
- ▌ Verification Before Completion · danmestasUse when about to claim work is complete, fixed, or passing, before checkpointing or requesting review — requires running verification commands and confirming output before making any success claims; evidence before assertions always
- ▌ Deterministic Simulation Testing · danmestasUse when building or testing distributed systems, consensus protocols, sync engines, replicated databases, or any system with network/disk/time non-determinism. Also use when tests are flaky due to concurrency, when debugging rare heisenbugs, or when asked about simulation testing, BUGGIFY, VOPR, or fault injection strategies.
- ▌ Frontend Component Decision Trees · danmestas bundleUse this skill whenever the user needs to choose the correct UI component for forms, notifications, errors, loading, buttons, overflow, onboarding, or design system contributions. Always start with the Main Entry Point dispatcher to select the right tree, then follow the exact branching logic. This skill prevents ad-hoc component choices by enforcing production design system decision trees.