Boruna Manage Tool
boruna_manageExecute, validate, and audit Boruna scripts with deterministic capability-safe runtime. Manage script execution, policies, and skill lifecycle.
Instructions
Boruna deterministic capability-safe .ax script runtime (v1.x LTS) — execute, validate, and audit Boruna scripts. Side effects depend entirely on the script's declared capabilities. Requires an active mcp_stdio Tool record pointing to a working Boruna binary; otherwise every action returns dependency_unavailable.
Actions:
run (write — side effects per script capabilities) — inline
.axsource ORskill_idof a saved boruna_script skill; optional: timeout_ms, mem_limit_mb.validate (read) —
.axsource. Syntax + semantic check, no execution.policy_validate (read; v0.4.0+) — policy JSON. Strict schema validation.
evidence (read) — run_id. Capability/effect evidence record for an executed run.
capability_list (read) — registered capabilities and capability_set_hash.
skill_manage (write) — sub-actions list/get/create/update/delete on boruna_script-typed skills.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| action | Yes | Action to perform: run, validate, policy_validate, evidence, capability_list, skill_manage | |
| 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. | |
| mode | Yes | Execution mode: inline (run script directly) or skill (run a saved boruna_script skill) | |
| script | No | (inline mode) The .ax script source code to execute | |
| policy | No | (inline mode) Legacy capability policy shorthand: "allow-all" or "deny-all" (default: deny-all). For Boruna v0.2.0+ fine-grained gating, use policy_structured instead. | |
| policy_structured | No | (inline mode, Boruna v0.2.0+) Structured Capability Policy object with required default_allow (bool), optional rules (per-capability {allow, budget}), and optional net_policy (allowed_domains, allowed_methods, max_response_bytes, timeout_ms, allow_redirects). Capability keys: net.fetch, fs.read, fs.write, db.query, ui.render, time.now, random, llm.call, actor.spawn, actor.send. When set, takes precedence over the legacy policy parameter. See https://github.com/escapeboy/boruna/blob/v0.2.0/docs/reference/policy-schema.md. | |
| boruna_tool_id | No | (inline mode) UUID of the mcp_stdio Tool pointing to the Boruna binary. If omitted, auto-detects. | |
| input | No | Optional input data passed to the script as JSON | |
| limits | No | (inline mode, v1.0+) Optional resource limits: max_wall_ms (int) and/or max_output_bytes (int) | |
| skill_id | No | (skill mode) UUID of the boruna_script Skill to execute | |
| policy_json | Yes | JSON-encoded Boruna policy object to validate. Must have a top-level default_allow boolean. Example: {"default_allow":false,"rules":{"net.fetch":{"allow":true}}} | |
| execution_id | Yes | UUID of the SkillExecution record from a boruna_script skill run | |
| name | No | (create) Skill name | |
| description | No | (create) Skill description | |
| limit | No | (list | executions) Max results (default 20) |