Cloud Skill Manage Tool
skill_manageCreate, manage, and execute reusable AI skills including LLM prompts, connectors, rules, and guardrails. Supports direct execution of code, browser, and multi-model tasks.
Instructions
AI skills — reusable units agents invoke (LLM prompt templates, connector calls, rules, hybrid pipelines, guardrails). Skill executions are versioned (SkillVersion) and metered against the team budget. Direct execution actions (guardrail, multi_model, code_exec, browser) bypass the agent layer for ad-hoc invocation by humans/assistants.
When to use: build reusable prompt or workflow primitives that multiple agents share, or run a one-off skill (e.g. code_exec to evaluate Python in a sandbox). Do NOT use for raw LLM calls — use assistant_manage instead.
CRUD actions:
list (read) — optional: type (llm|connector|rule|hybrid|guardrail), status, limit (default 50).
get (read) — skill_id. Returns full config + latest version.
create (write) — name, type, config (type-specific JSON). Returns skill_id.
update (write) — skill_id + any creatable field. Bumps version; old version retained.
delete (DESTRUCTIVE) — skill_id. Soft-deletes; existing version history retained for 30 days.
versions (read) — skill_id. Version log with diffs and created_by AiRun id.
Direct execution (each costs credits; rate-limited per skill):
guardrail (read — costs ~1-3 LLM credits) — input (string), rules (array). Returns { passed: bool, violations: [...] }.
multi_model (write — costs ~N × LLM credits where N=models[].length) — prompt, models[]. Runs same prompt across providers and returns consensus.
code_exec (DESTRUCTIVE — runs in sandboxed Docker, costs sandbox-minute credits) — code, language (python|node|bash). Network-disabled by default; max 60s wall-clock; max 256 MB memory.
browser (write — costs browser-minute credits) — url, actions (array of {type, selector, value}). Headless Chromium with SSRF guards.
Cloud note: supabase_edge_function not exposed in cloud (use integration_execute with a Supabase integration instead).
Errors: 401, 403, 404, 422, 429, 503 (sandbox/browser worker pool exhausted).
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| action | Yes | Action to perform: list, get, create, update, delete, versions, guardrail, multi_model, code_exec, browser | |
| deadline_ms | No | Optional: max wall-clock time (ms) the tool may spend. If exceeded during the call, returns a DEADLINE_EXCEEDED error. Minimum 100 ms. Leave unset for no deadline. | |
| type | No | Filter by type: llm, connector, rule, hybrid | |
| framework | No | Filter by framework key (e.g. rice, spin, bant, okrs, unit_economics). See framework_list for full enum. | |
| limit | No | Max results to return (default 10, max 100) | |
| skill_id | Yes | The skill UUID | |
| name | Yes | Skill name | |
| description | No | Skill description | |
| prompt_template | No | System prompt template for LLM-backed skills | |
| data_classification | No | Data classification level: public, internal, confidential, restricted. | |
| step_id | No | For get_result. The playbook step UUID. | |
| workflow_node_id | No | For set_node_guardrail / remove_node_guardrail. The workflow node UUID. | |
| guardrail_skill_id | No | For set_node_guardrail. The guardrail skill UUID to attach. | |
| execution_id | No | For get_execution: the SkillExecution UUID. | |
| worktree_execution_id | No | For get_execution / get_diff: the WorktreeExecution UUID. | |
| status | No | For list_executions: filter by status (pending_approval, completed, failed, approved, rejected). |