AI & ML

delorenj Bundle

Controlled self-referential learning loop with structured observations, external runtime state, Holyfields-defined learning events, and manual promotion/rollback of lessons.

delorenj Bundle

Use when building, editing, validating, testing, or debugging an n8n workflow through the n8n-mcp MCP server — designing a flow, configuring a node, writing an expression or Code node, wiring credentials, or fixing one that misbehaves. The entry-point skill for the n8n-mcp-skills pack routes you to the right specialist skill, gives working knowledge of every n8n-mcp tool from turn one, and states the rules that keep workflows from breaking in production. Always consult it first on any n8n, workflow, node, or automation task — even a quick one-off, and even when the user names no skill — because n8n's surface drifts between versions and the specialist skills prevent silent failures.

delorenj Bundle

Debug, configure, and extend delorenj's single-session zellij workspace and the agent surfaces bound to it. Use when touching files in ~/.config/zellij, when a zellij CLI call hangs or times out, when deckard or zellij-driver misbehave, when agent panes come back suspended after a crash, when wiring a zellij plugin or keybind, when tab attention/notification does not reach the user, or when an agent must find/focus/rename a tab from outside a pane. Triggers: zellij, config.kdl, load_plugins, MessagePlugin, zellij pipe, zellij action, override-layout, session-layout.kdl, resurrection, start_suspended, post_command_discovery_hook, agent-pane, zellij-doctor, unwedge, Workspace session, ZELLIJ_PANE_ID, deckard, DECKARD_PLUGIN, tab attention, visual bell, zjstatus, zellij plugin wasm. Do NOT use for tmux, for generic terminal-emulator config (alacritty/ghostty), for deckard's internal Rust architecture (read that repo), or for Bloodbank event schemas (bloodbank-integration).

kevinnft Skill

Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.

kevinnft Bundle

TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.

kevinnft Bundle

Build a throwaway prototype to flush out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".

delorenj Bundle

Use when a task needs an isolated hidden Linux desktop or workspace-owned browser: GUI app QA, web/browser/shopping automation, sandboxed app observation, or stale workspace cleanup. Routes agent-workspace-linux MCP tools on demand. Does NOT apply to host desktop/Chrome control, generic MCP setup, or pure code/file edits.

jacob-balslev Skill

Use when designing or evaluating a production AI agent system, choosing a multi-agent coordination pattern (orchestrator/worker, fan-out, consensus, sequential chain, evaluator/optimizer), diagnosing coordination failures (claim races, silent stalls, context contamination, runaway loops), or auditing whether an agent loop is truly production-ready. Covers the four pillars (architecture and lifecycle, task decomposition, coordination patterns, production reliability), the six reliability requirements (observability, cost budgets, idempotency, failure recovery, safety caps, claim locks), the delegation decision framework with overhead crossover, and the most common anti-patterns. Do NOT use for prompt wording (use `prompt-craft`), per-call tool efficiency (use `tool-call-strategy`), context-stack design within a single agent (use `context-engineering`), or runtime debugging of a deployed system (use `debugging`). Do NOT use for improve this prompt's wording to get better outputs.

jacob-balslev Skill

Use when deciding what to load into an active agent session, recovering from context drift, preparing compaction or restart, distilling raw inputs into a working summary, or writing a handoff another agent can resume quickly. Covers intake triage, the active-context contract, the six-step context-management loop, working-set shaping, evidence capsules and observation masking, just-in-time handle-before-payload loading, prompt-caching alignment, context-rot defenses, drift signals and reset, the runtime-primitives boundary, compaction-ready handoffs, and selective rebuild after context loss. Do NOT use for token math (use `context-window`), prompt wording (use `prompt-craft`), persistent memory curation, or multi-graph context architecture (use `context-graph`). Do NOT use for calculate the per-zone token budget for the 200K context window. Do NOT use for improve this prompt template for the grader. Do NOT use for curate the persistent memory index file.

kevinnft Bundle

Configure, extend, or contribute to Hermes Agent.

kevinnft Skill

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.

dykyi-roman Bundle

Generates Read Model/Projection for PHP 8.4. Creates optimized query models for CQRS read side with projections and denormalization. Includes unit tests.

dykyi-roman Bundle

Diagram knowledge base. Provides Mermaid syntax, C4 model, diagram types, and best practices for technical diagrams.

delorenj Bundle

Optimize multi-agent coding CLI stacks for lower token burn across Claude Code, Codex, Kimi Code CLI, and Hermes; includes workflows for unknown agents, cross-agent updates, MCP/profile propagation, and usage/provider rotation analysis.

delorenj Bundle

Operate the Skillex skill registry at ~/code/skillex: author, version, render, seal and verify PACKS; curate all-skills/ and skill-sets/; declare packs and skills in .agents/skills.json. Use when cutting or upgrading a pack, running skillex pack render/verify/manifest, editing a global or project .agents/skills.json, resolving packs[] vs skills[] precedence and redundant entries, wiring the mise skills-sync and skills-provision-packs tasks, or fixing pj audit failures on skills.project-manifest. Triggers: skillex, skill pack, packs[], pack.toml, SHA256SUMS, sealed pack, payload_files, all-skills, skill-sets, skills.schema.json, inherit_global, PJ_SKILLS_REGISTRY_ROOT, provision-packs.py, sync-skills.py, accept-registry-matches, bmad pack. Do NOT use for: SSOT fan-out engine mechanics or hooks.master.json (agent-config-fanout); SKILL.md content or topology (skill-creator); pjangler Commands/Recipes (project-jangler); project bootstrap (33god-projects); ~/.hermes/skills overlay (33god-agent-fleet-operations).

delorenj Bundle

Join a Google Meet or Zoom call as a video meeting agent via PikaStreaming. Trigger: user drops a Google Meet or Zoom link, or asks to join a meeting.

kevinnft Bundle

W&B: log ML experiments, sweeps, model registry, dashboards.

kevinnft Bundle

Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.

kevinnft Bundle

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

jacob-balslev Skill

Enforce five non-negotiable quality pillars as a pre-output gate: complete reporting (show ALL items, never filter unilaterally), verification (no claim of works/done/exists without a tool-call receipt in the same turn), thoroughness (every acceptance criterion verified with evidence; docs ship with the change), enrichment ('improve' adds capability, never trims), and anti-shortcut (exhaust deterministic lookup before guessing; findings demand action, not filing). Use when reviewing enumerated output for completeness, when an agent claims something works without evidence, when marking a task done, or when asked to 'improve' anything.

jacob-balslev Skill

Use when turning raw qualitative research output (interview transcripts, field notes, diary entries, observation logs) into themes, patterns, insight statements, and design-ready artifacts via affinity mapping, empathy maps, and jobs-to-be-done framing. Do NOT use for collecting new research, quantitative analysis, statistical inference, or summarizing a single document — synthesis specifically operates on a corpus of qualitative evidence. Do NOT use for Run more interviews with five additional participants. Do NOT use for Calculate the response rate of the survey. Do NOT use for Summarize this single PDF document. Do NOT use for transforming a corpus of collected qualitative evidence into themes and insights (use user-research).

jacob-balslev Skill

Use when an agent is making too many tool calls, when context is filling from verbose tool outputs, when the same operation could be a script instead of N individual calls, or when designing a tool-use protocol for a new agent or harness. Covers the three costs of every call (token, latency, context pollution), the script-vs-call decision gate, tool-selection decision trees (file-search vs content-search vs targeted-read vs full-read), call batching and parallelization, redundancy avoidance, the poka-yoke principle, subagent delegation for context protection, and cost-benchmark heuristics by task type. Do NOT use for prompt wording (use `prompt-craft`), broader context stack design across the five layers (use `context-engineering`), runtime tool failures or production debugging (use `debugging`), or behaviour-preserving refactor mechanics (use `refactor`). Do NOT use for improve this prompt's wording to get better outputs. Do NOT use for design what skills get loaded for which prompts.

jacob-balslev Skill

Use when designing what information reaches an LLM agent before it reasons — system prompt, persistent memory, always-loaded rules, injected skills, and the user prompt — or when diagnosing why an agent produced a wrong answer despite a clear instruction. Covers the four context failure modes (missing, stale, wrong, overwhelming), the five-layer context stack, four context quality metrics (injection precision and recall, utilization, freshness), the Frequent Intentional Compaction (FIC) protocol, subagent delegation for context-heavy work, and the failure-mode decision tree. Do NOT use for prompt wording (use `prompt-craft`), authoring a new SKILL.md (use `skill-scaffold`), or deciding which skill the router activates for a given query (use `skill-router`). Do NOT use for improve this prompt's wording to get better outputs. Do NOT use for scaffold a new SKILL.md for our team's deploy procedure. Do NOT use for the router picked the wrong skill for this query — debug it.

jacob-balslev Skill

Use when authoring any artifact that makes claims — skill content, documentation, audit findings, architecture proposals, code review comments, research summaries, or agent output. Covers the discipline of grounding every claim to a verifiable source, distinguishing verified-by-evidence from inferred-from-context, Chain-of-Verification-style claim checking, RFC 2119/RFC 8174 modality, Toulmin argument structure, citation faithfulness, native citation-tool limits, and honest handling of inference, absence, conflict, and currentness. Do NOT use for execution-level evidence protocols or output-completeness enforcement (use methodology), naming precision (use semantics), or grader/rubric design (use evaluation or eval-driven-development). Do NOT use for decide which lint rule to add for a specific kind of drift (use skill-infrastructure). Do NOT use for evaluate a finished SKILL.md against the comprehension grader (use evaluation).

delorenj Bundle

Build voice agents with the Cartesia Line SDK. Supports 100+ LLM providers via LiteLLM with tool calling, multi-agent handoffs, and real-time interruption handling.

delorenj Bundle

Excavates lost feature context from git history, docs, tickets, PRs, and session breadcrumbs to restore missing project memory to Hindsight. Use when the user says a feature has regressed/vanished/been forgotten, summons GitArcheologist, or asks to dig up the remains of past work.

jscraik Bundle

Best practices for working with Go codebases. Use when writing, debugging, or exploring Go code, including reading dependency sources and documentation.

jscraik Bundle

Create and review TOML configuration with strict typing and predictable structure. Use when editing tool configuration files that require schema-safe TOML.

jscraik Bundle

Install, repair, and validate Vale prose linting. Use when users need Vale config, style sync, docs lint gates, or broken Vale workflow diagnosis.

jscraik Bundle

Create and review YAML files with safe indentation, schema-aware structure, and low-surprise serialization. Use when editing YAML config or workflow files.

jscraik Bundle

Review file-based todo findings into ready, skipped, customized, or blocked states. Use this skill when pending todo files need approval.

jscraik Bundle

Apply approved fixes for unresolved CodeRabbit review comments, Codex P1-P3 findings, PR feedback, and code review issues with validation evidence. Use when asked to address review comments, fix review findings, clear unresolved comments, or autofix PR feedback.

jscraik Bundle

Validate and choose proportionate test proof for tests, CI, coverage, evals, and closeout evidence: map changed files to repo-native commands, place checks at commit, push, and pull-request gates, classify pass/fail/blocked ownership, and preserve trace/regression artifacts. Use when users ask what tests or gates to run, how to design coverage, why validation failed, or what proof supports a claim.

kevinnft Skill

Tests in real browsers. Use when building or debugging anything that runs in a browser. Use when you need to inspect the DOM, capture console errors, analyze network requests, profile performance, or verify visual output with real runtime data via Chrome DevTools MCP.

kevinnft Skill

Extract a DDD-style ubiquitous language glossary from the current conversation, flagging ambiguities and proposing canonical terms. Saves to UBIQUITOUS_LANGUAGE.md. Use when user wants to define domain terms, build a glossary, harden terminology, create a ubiquitous language, or mentions "domain model" or "DDD".

jscraik Bundle

Analyze current external library or API docs with Context7 when dependency behavior, version-sensitive references, or ctx7 CLI setup/install guidance is needed.