@openagentid/skills-policy (TypeScript)
Governance for agent skills — a specialized Arsenal policy that gates which SKILLS.md entries an agent may invoke, with rate limiting, argument constraints, time windows, consent, and hash-chained audit receipts.
Purpose
Governance for agent skills — a specialized Arsenal policy that gates which SKILLS.md entries an agent may invoke, with rate limiting, argument constraints, time windows, consent, and hash-chained audit receipts.
Source contract
| Field | Value |
|---|---|
| Package | @openagentid/skills-policy |
| Version | 0.1.1 |
| Language | TypeScript |
| Runtime floor | {} |
| Manifest | openagent-sdk/crates/openagent-skills-policy/typescript/package.json |
| Distribution | Source package; registry publication not verified |
Manifest SHA-256: 722a94f6f21a20d6b2eeb922f09468483f466028d415fd53b61c27fd29d68d64. The recorded Git revision identifies the containing repository; the manifest and entry-point hashes identify this working-tree snapshot.
Integration
Use the package root above from an authorized source checkout. Preserve sibling dependencies and their version/feature requirements. Install the runtime and compile the selected package before using its generated artifacts. The task guide describes configuration, trust boundaries, and error behavior. Test/conformance and deployment packages are supporting tools, not interchangeable production identity providers.
Declared package subpaths
{
".": {
"types": "./dist/index.d.ts",
"import": "./dist/index.js"
}
}Exported entry points
The declarations below are extracted from the named entry files. This is a navigable source reference, not compiler-generated documentation of every transitive module. Feature gates and runtime requirements still apply. Private implementation files are not promoted to public APIs merely because they contain a public declaration.
openagent-sdk/crates/openagent-skills-policy/typescript/src/index.ts
Source SHA-256: c1907919466729c24cc00553a48156bd77aa6c6d36b77b9a695ab38880e5395f.
export { AuditChain, type Receipt, HASH_LEN } from "./audit.js";
export { Did } from "./did.js";
export { InvocationContext, SkillsPolicy } from "./engine.js";
export {
ArgumentConstraintError,
ConsentRequiredError,
InvalidPolicyError,
InvalidSkillsMarkdownError,
NotAllowedError,
OutsideTimeWindowError,
RateLimitExceededError,
SkillsPolicyError,
} from "./errors.js";
export { RateLimiter, type RateLimitDecision } from "./rate.js";
export {
type AuditLevel,
CURRENT_VERSION,
type DefaultRule,
MAX_SKILL_RULES,
type RateLimit,
type SkillRule,
type SkillsPolicyDoc,
type TimeWindow,
} from "./schema.js";
export { type SkillEntry, SkillsManifest } from "./skillsMd.js";Exported API signatures
openagent-skills-policy (Rust)
Governance policy engine for agent skills — an Arsenal-composed policy layer that gates which SKILLS.md entries an agent may invoke, with rate limiting, argument constraints, time windows, consent, and hash-chained audit receipts.
openagent-aws-lambda (TypeScript)
Deployment adapters source package and exported entry points.