Use when the user asks a graph or tabular question with an exact answer (reachability, transitive closure, shortest or weighted paths, counts, aggregations, set operations, joins), or mentions connectors, external tools, invoke, affordances, or "which tool". Route exact symbolic work to computeoffload.routeoperation before reasoning, and keep connector spokes behind toolsearch and invoke.
EnglishTeach and run Theorem's Harness cross-agent coordination protocol. Use when Codex, Claude Code, Claude.ai, or another head works the same repo/task and the user wants coordination, shared intent, mentions, presence, handoff, or a ping-like wait. The model is one agent with several heads sharing a scratchpad, announcing over a shared room and checking semantic overlap before reconciling concrete edits.
EnglishThe Harness planning capability, backed by the durable plan substrate. Use when the task needs stable acceptance criteria, a multi-session plan, a migration shape, or a plan worth handing to another agent. Reachable as /harness mode=plan or, for users who explicitly want planning, the /planning-theorem compatibility command.
EnglishShow the event timeline of a Theorem's Harness run or the bounded replay of a plan. Use when the user asks "what did you actually do," "show me the run trace," "walk me through the run," "audit what happened," "what tools fired and in what order," "why was this task refused," or "what happened on this plan." Wraps the harnessrun MCP tool for run timelines and the harnessreplay MCP tool for plan-scoped transition/refusal replay.
EnglishUse when the user asks to "write Rust", "debug Rust", "review Rust", "fix Cargo", "work on a Rust MCP server", "improve a Rust crate", or needs help with Cargo workspaces, Rust validators, PyO3/FFI bridges, async Rust services, parsers/macros, systems code, Ensemble pack selection, affordance charters, or Rust skill-pack corpus work.
EnglishTheorem's Harness default plugin skill. Use when the user invokes /harness or asks for grounded planning, implementation, debugging, review, context preparation, validation, reporting, cross-agent coordination, typed memory, encode, run lifecycle, replay, fork/compare, fractal search, code search, or harness-backed agent work.
English
AGENT SKILLS ATLASIncludes Affordance Router, Harness Coordinate, Planning Theorem, and other workflows.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLISupported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Affordance Router
Use when the user asks a graph or tabular question with an exact answer (reachability, transitive closure, shortest or weighted paths, counts, aggregations, set operations, joins), or mentions connectors, external tools, invoke, affordances, or "which tool". Route exact symbolic work to computeoffload.routeoperation before reasoning, and keep connector spokes behind toolsearch and invoke.
Harness Coordinate
Teach and run Theorem's Harness cross-agent coordination protocol. Use when Codex, Claude Code, Claude.ai, or another head works the same repo/task and the user wants coordination, shared intent, mentions, presence, handoff, or a ping-like wait. The model is one agent with several heads sharing a scratchpad, announcing over a shared room and checking semantic overlap before reconciling concrete edits.
Planning Theorem
The Harness planning capability, backed by the durable plan substrate. Use when the task needs stable acceptance criteria, a multi-session plan, a migration shape, or a plan worth handing to another agent. Reachable as /harness mode=plan or, for users who explicitly want planning, the /planning-theorem compatibility command.
Replay Last Run
Show the event timeline of a Theorem's Harness run or the bounded replay of a plan. Use when the user asks "what did you actually do," "show me the run trace," "walk me through the run," "audit what happened," "what tools fired and in what order," "why was this task refused," or "what happened on this plan." Wraps the harnessrun MCP tool for run timelines and the harnessreplay MCP tool for plan-scoped transition/refusal replay.
Rust Engineering
Use when the user asks to "write Rust", "debug Rust", "review Rust", "fix Cargo", "work on a Rust MCP server", "improve a Rust crate", or needs help with Cargo workspaces, Rust validators, PyO3/FFI bridges, async Rust services, parsers/macros, systems code, Ensemble pack selection, affordance charters, or Rust skill-pack corpus work.
Theorems Harness
Theorem's Harness default plugin skill. Use when the user invokes /harness or asks for grounded planning, implementation, debugging, review, context preparation, validation, reporting, cross-agent coordination, typed memory, encode, run lifecycle, replay, fork/compare, fractal search, code search, or harness-backed agent work.
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Affordance Router skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- Reported license for the skills shown: MIT. The author’s terms apply.