菜单
Create, inspect, or modify Spell model profiles and agent profile files. Use when the user asks to configure providers, add or change `.agent.edn` files, expose namespaces, define sub-agents, adjust budgets, or use `spell.api/run`.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Navigate and develop the Spell source code. Use when modifying Spell internals, investigating evaluator/runtime/provider behavior, finding the right source files, updating tests, or explaining how the implementation fits together.
Set up Spell from a fresh checkout and run the first smoke tests or examples. Use when a user asks to install Spell, get started, verify the CLI, configure provider credentials, choose a first example, or try Spell for the first time.
Use Spell from the CLI, the Clojure API, and the Python benchmark adapter. Use when writing commands, embedding Spell in Clojure, calling Spell from benchmark Python code, or checking current API vocabulary.
Configure custom Spell agent and model profiles. Use when creating or modifying .agent.edn files, choosing namespaces, changing system prompts, wiring sub-agents, selecting provider-prefixed models, or doing agent smoke tests.
Navigate and develop the Spell source code. Use when modifying Spell internals, investigating evaluator/runtime/provider behavior, finding the right source files, updating tests, or explaining how the implementation fits together.
Set up Spell from a fresh checkout and verify prerequisites, CLI smoke tests, and available model provider authorization. Use when helping a user get to a runnable checkout, configure API provider access, or try the first Spell example.
| name | spell-agent-config |
| description | Create, inspect, or modify Spell model profiles and agent profile files. Use when the user asks to configure providers, add or change `.agent.edn` files, expose namespaces, define sub-agents, adjust budgets, or use `spell.api/run`. |
Use this skill for runtime configuration, not for changing evaluator semantics.
Keep docs/api.md as the public API and configuration reference. It is the right home for stable docs because it is user-facing, linkable from the README, and useful outside agent environments. This skill should summarize workflow and point to it, not replace it.
Read before editing config:
sed -n '1,320p' docs/api.md
sed -n '1,180p' config/AGENTS.md
Model profiles live in config/model-profiles/*.edn. They describe how Spell calls a model provider: provider type, endpoint/auth, default model, reasoning effort, timeouts, retry policy, pricing, and transport options.
Agent profiles live in config/agent-profiles/*.agent.edn. They describe what the model sees and can do: system prompt, namespaces, sub-agent topology, recovery behavior, structured output, and default budget.
Run-level overrides belong in spell.api/run or CLI flags: task input, model override, reasoning effort override, budget, depth, trace directory, usage tracker, user reader, and log writer.
When creating or editing an agent profile:
base-pf.agent.edn, base-msg.agent.edn, or base-tc.agent.edn.:agent-name, :agent-description, :system-prompt, :default-model-profile, :default-budget, :available-agents, :namespaces.stdlib/io, stdlib/web, stdlib/patterns, stdlib/agents, and stdlib/globals.:available-agents for worker discovery instead of exposing .agent.edn files as namespace values.When creating or editing a model profile:
:provider supported by spell.provider/resolve-model-profile.:default-model and :default-reasoning-effort for model-call defaults.:api-key-env, not literal secrets, for public config.:default-agent-profile transport-compatible with the provider.data/pricing.edn when adding a priced model.Do not add backwards-compatibility aliases or legacy config shims unless explicitly requested.
Run the smallest useful checks:
bin/spell -h
bin/spell -t "Return 42"
clj -M:test-fast
For API changes, add focused tests under test/spell/api_test.clj, test/spell/agent_test.clj, or provider-specific tests.