AI & ML

ashermahonin Bundle

Keep a repository understandable while an agent creates files, tests, scripts, documentation, fixtures, screenshots, migrations, or generated output. Use during project setup, implementation, review, release, and documentation work. Maintain a clear project map, isolated temporary artifacts, one canonical documentation entrypoint, and a cleanup check before handoff.

android Bundle

Analyzes Android codebases to implement ML Kit GenAI Prompt API. Use this skill to send natural language requests on-device to Gemini Nano, use structured output with Prompt API, implement prefix caching, optimize the current prompt, or apply best practices."

yangliu2060 Bundle

STORM 调研助手：用多视角证据分析选题、产品、工具或工作流是否值得推进。

alibaba-skill-up Bundle

Verify Sandbox MCP can be installed and called from skill-up.

alibaba-skill-up Bundle

Per-case mocked MCP override smoke skill used by skill-up e2e tests.

alibaba-skill-up Bundle

Reviews code and provides detailed quality assessment

qodo-ai Bundle

Connect Qodo to the current local coding agent — verify the Qodo CLI, guide a secure installation when it is missing, complete browser login, and confirm managed tools are ready. Use after installing the Qodo plugin, when the user asks to set up or connect Qodo, or when another Qodo skill reports that the CLI is missing or logged out.

qodo-ai Bundle

Connect Qodo to the current local coding agent — verify the Qodo CLI, guide a secure installation when it is missing, complete browser login, and confirm managed tools are ready. Use after installing the Qodo plugin, when the user asks to set up or connect Qodo, or when another Qodo skill reports that the CLI is missing or logged out.

randommonicle Skill

Before declaring any deployed or user-facing change done, shipped, or closed out, run one real end-to-end invocation through the real seams — the deployed function, a real login and JWT, the real model call, the real browser or device. Green CI, green smokes, and a clean typecheck prove the layers; only the ride proves the seams. Triggers on "done", "shipped", "live", closing out a feature, and after any migration that existing code reads. Does not fire mid-implementation on individual command results (verify-the-effect's norm covers per-signal checks), nor on changes with no deployed or user-facing surface.

randommonicle Bundle

Keep the context window small and its prefix stable, because every token in it is re-sent on every later turn. Read slices instead of whole files, cap tool and log output before it lands, isolate high-volume work behind a summary, and clear rather than compact between unrelated tasks. Triggers when about to read a large file, log, dataset, transcript or search result into context; when a session has grown long, auto-compacted, or hit a usage limit mid-task; when writing or growing an instruction file (CLAUDE.md, AGENTS.md, GEMINI.md) or a skill; and when assembling a prompt from a static part and a varying part. Does not fire on pricing a recurring job or reporting what an action cost (price-the-spend), shaping a multi-agent roster (commission-the-roster), writing the handover at the end of a session (handover), or designing a product path that sends data to a model (ai-surface-discipline).

randommonicle Bundle

Adversarially review a specific change, design, or finding by having Claude debate one or more INDEPENDENT AI agents (Gemini via the agy CLI or GPT via the codex CLI, driven by Claude with no pasting; or a chat the user drives; several seats at once) over a shared file relay, with mandatory file:line citations, live read-only evidence, and a converge-or-two-positions outcome. Use when the user wants a different model's adversarial take before acting, mentions Antigravity / cross-agent / agent-exchange / "hash it out with Gemini", or wants a finding stress-tested by independent models. NOT for whole-project audits (that is committee-review) or quick single-pass checks. Needs at least one second agent, driven headlessly through the CLI transport or by the user in a chat, and a shared directory both sides can read and write.

randommonicle Bundle

Stops fabricated and drifted references in any document that cites code, schema, config or history. Load BEFORE writing or reviewing a handover, design doc, plan, spec, decision log, lessons entry, audit, review report, PR description, or commit message that points at files, line numbers, migrations, commits, PRs or dates. The failure it prevents is specific and measured, a model grounds technical facts correctly while inventing the PATH around a borrowed line number, drifting a line number onto a blank line, citing a range that spans unrelated blocks, or misattributing when something landed, and then self-reports that everything was verified. Triggers on file:line, "landed in PR", "added on", "see some/file.ts", migration numbers, or any claim a reader would act on without re-deriving. Do not load for writing code itself.

randommonicle Skill

Playbook for the always-on norm "filters and gates hide empty things; they never drop content". Load when adding any filter, gate, continue, catch, or fallback to a rendering or export path, when parsing model output that carries content, or when two renderers must produce the same content. Does not fire on error-message wording or telemetry levels (honest-failure-surfacing) or on filters over data the user never receives.

randommonicle Skill

Before spawning a fleet of agents, publish a roster for the operator's approval - role, model tier, token budget, scope and single artifact per agent - and brief each agent with pre-verified facts instead of questions. Triggers when appointing a lead agent, delegating a multi-agent wave, planning a commission whose shape is not yet known, choosing which model tier does which job, or when a usage window emptied faster than the work justified. Does not fire on spawning one agent for one bounded task, and does not price the result (price-the-spend reports the burn afterwards).

randommonicle Skill

A subagent, reviewer, or scout report is evidence, not findings — and so is your own first-hand defect analysis. Re-derive every Critical or High claim, everything touching a regulatory anchor, and the reasoning about the fix from the primary source before acting or writing it into a durable artifact. Triggers when consuming any agent or reviewer output and when recording a defect or fix analysis. Does not fire on low-severity suggestions with no gating or regulatory consequence — sample those rather than re-deriving them all.

randommonicle Skill

When creating or reviewing any endpoint or control that touches money or shared state, two rules apply together. Every endpoint that spends money (LLM, email, SMS, payment) or writes shared state (bookings, slots) gets a per-IP volume cap the day it is created — anyone can POST directly, bypassing the UI. And any guardrail on that path written to fail open (origin allowlist defaulting to permit, auth that skips on a missing header) is inert until explicitly closed, and its failure is silent — name the real primary defence. Does not fire on internal-only endpoints behind service-role auth with no paid upstream; internal fan-out caps belong to lock-at-the-chokepoint.

nwave-ai Skill

Backward-compatible router for creating agents, validating existing agents, migrating existing agent monoliths, or optimizing one existing skill.

nwave-ai Skill

Configure explicit Codex or Claude model selections for nWave DES defaults or published roles. Use when choosing provider/model defaults or project overrides; it never derives a model from a competence label.

cklxx Bundle

When a subtask should run on a different agent (Codex/Claude/Gemini) → delegate and track cross-agent execution.

cklxx Bundle

When the user wants to interact with Moltbook (AI agent social network) → post, browse, comment, vote, search via shell curl.

cklxx Bundle

When agent config parameters need querying or modification → read/write YAML configuration files.

vinvcn Bundle

把当前对话交接给一个全新的 background agent，让它立即接手工作。

pskoett-pskoett-ai-skills Skill

Monitors context window health throughout a session and rides peak context quality for maximum output fidelity. Activates automatically after plan-interview and intent-framed-agent. Stays active through execution and hands off cleanly to simplify-and-harden and self-improvement when the wave completes naturally or exits via handoff. Use this skill whenever a multi-step agent task is underway and session continuity or context drift is a concern. Especially important for long-running tasks, complex refactors, or any work where degraded context would silently corrupt the output. Trigger even if the user doesn't say "context surfing" — if an agent task is running across multiple steps with intent and a plan already established, this skill is live.

pskoett-pskoett-ai-skills Bundle

Captures learnings, errors, corrections, and feature requests to enable continuous improvement. Use when: (1) User corrects Claude ('No, that's wrong...', 'Actually...'), (2) User requests a capability that doesn't exist, (3) Claude realizes its knowledge is outdated or incorrect, (4) A better approach is discovered for a recurring task, (5) Receiving a Handoff block from self-healing (a recurring verified heal with Recurrence-Count at least 3) to distill into a memory file or new skill. For ACTIVE runtime failures where the agent needs to apply and verify a fix mid-task, use `self-healing` instead (it files HEAL- entries with proof; self-improvement promotes accumulated patterns). Also review learnings before major tasks. For CI-only/headless learning capture, use self-improvement-ci.

pskoett-pskoett-ai-skills Skill

Frames coding-agent work sessions with explicit intent capture and drift monitoring. Use when a session transitions from planning/Q&A to implementation for coding tasks, refactors, feature builds, bug fixes, or other multi-step execution where scope drift is a risk.

pskoett-pskoett-ai-skills Skill

Frames coding-agent work sessions with explicit intent capture and drift monitoring. Use when a session transitions from planning/Q&A to implementation for coding tasks, refactors, feature builds, bug fixes, or other multi-step execution where scope drift is a risk.

pskoett-pskoett-ai-skills Skill

Control-plane workflow for coordinating multi-agent, multi-session project work from a single Codex, GitHub Copilot, or agent-app control session. Use this skill whenever the user asks to orchestrate agents, create or steer worker sessions, run a workflow-like effort, fan out audits/research/migrations, coordinate parallel implementation streams, monitor other project sessions, or compare this control-session pattern to Claude Code dynamic workflows. This skill is especially relevant when the current session can spawn persistent project sessions and those sessions can spawn their own subagents, creating a two-level orchestration hierarchy.

anthropics-claude-for-legal Bundle

Check regulatory feeds now and report what's new since the last check, filtered by your materiality threshold. Use when the user says "check the feeds", "what's new", "regulatory update", when running from the scheduled agent, or when manually pasting a regulatory development for classification and diff.

buldee Skill

Parallel agent orchestration for independent tasks. Use when facing 2+ independent tasks that operate on different files/domains and can be parallelized for efficiency.

stbenjam Bundle

Revisit superficial or under-verified agent work by investigating its claims, assumptions, evidence, and missing edge cases.

stbenjam Bundle

Stop punting and complete the user's requested task. Use after the model evades, defers, or stops before doing work it can perform.

ayoubben18 Skill

Produce a ready-to-run prompt for an autonomous /goal loop. Use when the user wants to hand off one continuous objective with a verifiable stop condition to run autonomously.

nwave-ai Skill

5-layer testing approach for agent validation including adversarial testing, security validation, and prompt injection resistance

nwave-ai Skill

Detailed review process, v2 validation checklist, and scoring methodology for agent definition reviews

nwave-ai Skill

Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation

nwave-ai Skill

Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation