智能体
Agent 规划、多步流程与编排
capability-evolver
EvoMap
A self-evolution engine for AI agents. Analyzes runtime history to identify improvements and applies protocol-constrained evolution. Communicates with EvoMap Hub via local Proxy mailbox.
find-reviewable-pr
dotnet
Finds open PRs in the dotnet/maui and dotnet/docs-maui repositories that are good candidates for review, prioritizing by milestone, priority labels, partner/community status.
buzz-guildhall-reporting
BryanHarrisScripts
Operational reporting procedure for summarizing and routing PlotPickle agent activity into appropriate BUZZ Guildhall rooms.
marquee-director
BryanHarrisScripts
Act as PlotPickle's private Key Art & Trailer Director after Foundations, using a narrow PPF-derived Marketing Context while the host controls generation and Marketing Reference storage.
create-app
atrislabs
Build and deploy an Atris app from a natural language description. Use when users ask to create Atris apps, workflows, or chat apps with setup automation.
chief-of-staff
atrislabs
"Daily briefing agent that learns your patterns and gets better every morning. Requires member-runtime skill. Use when you want a personalized morning briefing, daily summary, or 'brief me'."
symbiont
ThirdKeyAI
AI-native agent runtime with typestate-enforced ORGA reasoning loop, Cedar policy authorization, CommunicationPolicyGate for inter-agent governance, ToolClad declarative tool contracts, knowledge bridge, zero-trust security, multi-tier sandboxing, webhook verification, markdown memory, skill scanning, metrics, scheduling, symbi init/run/up/shell/repl CLI, interactive TUI (Beta), cross-instance agent messaging, human approval relay, and a declarative DSL
agent-customization
microsoft
'WORKFLOW SKILL — Create, update, review, fix, or debug VS Code agent customization files (.instructions.md, .prompt.md, .agent.md, SKILL.md, copilot-instructions.md, AGENTS.md). USE FOR: saving coding preferences; troubleshooting why instructions/skills/agents are ignored or not invoked; configuring applyTo patterns; defining tool restrictions; creating custom agent modes or specialized workflows; packaging domain knowledge; fixing YAML frontmatter syntax. DO NOT USE FOR: general coding questions (use default agent); runtime debugging or error diagnosis; MCP server configuration (use MCP docs directly); VS Code extension development. INVOKES: file system tools (read/write customization files), ask-questions tool (interview user for requirements), subagents for codebase exploration. FOR SINGLE OPERATIONS: For quick YAML frontmatter fixes or creating a single file from a known pattern, edit the file directly — no skill needed.'
Rafter -- Security Toolkit for AI Coding Agents
Raftersecurity
```
consistency-enforcement
akaszubski
Documentation consistency enforcement - prevents drift between README.md and actual codebase state. Auto-activates when updating docs, committing changes, or working with skills/agents/commands.
gitmoot
jerryfane
Use Gitmoot for local-first AI agent coordination across repositories, goals, reviews, PR comments, daemon jobs, stuck jobs, branch locks, agent templates, template capture and publish/pull, custom prompt agents, orchestration, heartbeats, event webhooks, the web dashboard, and Codex, Claude Code, or Kimi Code runtime workflows.
Observe Task For Learning
openvibely
Learn from one completed task and update durable skill configuration when useful.
evolve
evolving-machines-lab
"Evolve SDK development for TypeScript and Python. Use when building applications with Evolve to run AI agents (Claude, Codex, Gemini, Qwen, Kimi, OpenCode, Droid) in secure sandboxes. Triggers: (1) Creating Evolve applications, (2) Configuring agents with skills, Integrations, MCP servers, (3) Using Swarm abstractions (map, filter, reduce, bestOf/best_of, verify), (4) Building Pipelines, (5) Structured output with schemas, (6) Session management, streaming, observability, (7) Checkpointing, storage & StorageClient, (8) Cost tracking (per-run and per-session spend), (9) Historical sessions & trace download via sessions() client."
diagram
CxElegy
画流程图和架构图（中文、图为主、标清时间 / 费用 / 运行环境）。用户说「画流程图」「画架构图」「把 workflow 可视化」「哪个任务在哪个环境」「更新流程图」时使用。
hermes-agent-provider-switch
joecastelino
Diagnose why an AMG agent (Walter II, Don, Ralph, Stacey, etc.) is dead/unresponsive on an LLM-billing error, and switch it from OpenRouter to the ChatGPT-Team Codex OAuth provider. Use when an agent boots but can't reply, an ask-agent ping returns an HTTP 402 (out of credits) or 400 (unsupported model), or OpenRouter credits are exhausted. Covers the read_file API-key-masking landmine and how to discover the account's valid Codex model names.
atlas-event-driven-weights
kaecer68
"Use when modifying factor weights, adding factor types, or connecting narrative events to portfolio factor calculations. Triggers: factor weight logic changes, new factor type, FactorBridge, FactorWeightEngine modifications."
atlas-llm-provider-capability
kaecer68
"SOP for adding a new LLM provider client or capability handler in atlas-go. Covers BaseClient embedding, ProviderImpl interface, capability constant registration, routing table updates, handler wiring, and required tests. Trigger: user asks to add a new LLM provider, capability, or routing chain entry."
atlas-risk-status
kaecer68
"Interpret current risk status for retail investors. Explains the 4-tier risk architecture, VaR, drawdown protection, and risk gates with color-coded actionable guidance. Triggers: risk query, market entry timing, position reduction, 風險評估, 可以進場嗎."
sd-red-team
platypeeps
Use when the user wants a constructive adversarial review of an artifact's assumptions, contrary evidence, incentives, failure modes, misuse, security, privacy, counterarguments, and reversal conditions.
sd-objective-plan
platypeeps
Use when the user has accepted a goal or decision and wants a bounded, evidence-aware plan with observable milestones, dependencies, risks, decision points, and immediate next actions.
hermes-gateway-timeout-restart-loop
joecastelino
Diagnose and fix an AMG agent (Walter II/base or any profile) that is restarting/flaky with systemd "Failed with result 'timeout'" + status=9/KILL in the journal. Root cause is a hung LLM turn (e.g. gpt-5.5/Codex stall) holding an in-flight 300s API call that blocks clean shutdown past the 60s TimeoutStopSec, so systemd SIGKILLs the gateway and loops. This is NOT a crash, NOT out-of-credits, and the OpenRouter fallback can't save it. Covers how to tell this apart from a billing failure, the exact systemd-log signature, and the per-agent (and fleet-wide) drop-in fix.
dash-research-vault
Fearvox
"Multi-agent research vault with Evensong-style memory causation, Ebbinghaus decay, and L2 pressure self-evolution. Three-tier memory (public/restricted/internal), trigger-based agent activation, and 72-hour difficulty-weighted decay algorithm. For research teams building collaborative agent systems with structured long-term memory."
hmasd-jev-pro-transport
CartmanFatass
On the WSL host, directly send one committed HMASD Pro question through Jev Ultrafast and local Chrome, then use deterministic detached observation to collect completion and read the assigned repository target or preserve chat text. Never science, never a resend.
hmasd-chatgpt-pro-transport
CartmanFatass
In the current authoring session, send one committed HMASD Pro question through Agentify, arm deterministic observation, read the complete answer at the assigned repository target or preserve a complete chat fallback, and close the tab. Never science, never a resend.