メニュー
このリポジトリの skills
SkillsMP は arturcrmbot/zava-control-plane から 229 件の skill を収集しています。skill を開くとソースと詳細を確認できます。
arturcrmbot/zava-control-plane収集済み skill 229 件中 29 件を表示しています。
Score an IT access request as low, medium or high risk by combining the requester's recent breach history, the audit ledger's recent grant volume for the requested entitlements, and the permission depth of each selected role template.
原文の言語: 英語
Enumerate the role templates referenced in an IT access request, fetch each template's permissions, and screen the union against the employee's grade-band defaults for separation-of-duties conflicts.
原文の言語: 英語
Monitors the fleet of concurrent workflows across every registered domain (api/shared/domains.py). The FastAPI session appends a per-domain catalogue at boot so this skill stays domain-neutral. Composes the exception queue, amplifies operator skill by…
原文の言語: 英語
Draft a proposed performance rating + narrative for the reviewee by combining the cycle's OKR results with the grade-band distribution norm and the reviewee's calibration history.
原文の言語: 英語
Aggregate 360-degree peer reviews for the cycle, re-confirm the reviewee's reporting line via Workday HR, and pull the cycle's OKR results from the feedback collector.
原文の言語: 英語
Determine whether a proposed trip is in-policy and which cost band it lands in (low / mid / high) given the available booking options.
原文の言語: 英語
Look the proposed vendor up in the registry, list their regulatory filings, and screen the legal entity against sanctions for the country of incorporation and any country the filings reference.
原文の言語: 英語
Enumerate the vendor's ultimate beneficial owners, screen each one against sanctions, and run an adverse-media sweep on the top three by ownership percentage.
原文の言語: 英語
Coordinate the interview panel — find a slot across 4-5 panelists across timezones, send invites, gather feedback. RSVP collection is a HITL wait at the activity level.
原文の言語: 英語
Given a candidate's CV profile, screening verdict, voice transcript and role context, recommend whether to advance them (to interview at gate 1, or to offer at gate 2) and at what level. Recommends only — never decides.
原文の言語: 英語
Draft a job description from a req-to-hire payload, applying jurisdiction-appropriate boilerplate (USA vs Germany BetrVG works-council language) and the agency's voice guidelines.
原文の言語: 英語
Route the candidate's compliance check to the jurisdiction-appropriate sub-skill (USA → standard EEO checks; Germany → BetrVG works-council notification + DE labour law). Drives the §4.10 demo where flipping the country flag adds a Compliance step without…
原文の言語: 英語
Parse line items from a multi-line invoice payload.
原文の言語: 英語
Compose a breach-notification payload — Adaptive Card body for Teams + plain-text email body — for a Red expense claim. Cite the policy clause verbatim and request justification within 72 hours.
原文の言語: 英語
Compose the personalised offer letter from the negotiated terms and the JD's compensation framework. Surface it for HR BP HITL approval before the non-revocable send (gated by an onPreToolUse hook per §4.13).
原文の言語: 英語
Orchestrate the new hire's day-1: ServiceNow JML provisioning (laptop + accounts + access groups), Azure AI Speech avatar welcome video, Microsoft Graph day-1 calendar invites. Hook-gated for the JML send. Per spec §4.5 + §4.13.
原文の言語: 英語
Classify expense claim lines as Red/Amber/Green against the Zava synthetic T&E policy, citing the literal policy clause and exposing competing interpretations for boundary cases.
原文の言語: 英語
Cross-validate an expense claim's receipt image against the claim's structured fields (vendor, amount, date, line item). Detect six mismatch flavours: correct, wrong-amount, wrong-date, wrong-vendor, missing-line-item, missing-receipt.
原文の言語: 英語
Recommend an action for a classified reconciliation exception.
原文の言語: 英語
Propose a root cause explanation for an unmatched bank statement item.
原文の言語: 英語
Post the JD to Greenhouse and run a parallel LinkedIn search for passive candidates. Aggregate the inbound + outbound pipeline into a single candidate pool and hand it to Triage.
原文の言語: 英語
How to OCR a document via Azure AI Document Intelligence — when to use which prebuilt model, how to call ocr_extract, how to interpret confidence scores, error handling.
原文の言語: 英語
Conduct an inbound voice screen with a candidate via ACS / GPT-Realtime. Ask 4 calibrated rubric questions, abort on red-flag responses, score the transcript, return a shortlist signal. Per spec §4.5.
原文の言語: 英語
Sub-skill of `compose-domain` (NEW v3). Writes ONE persona SKILL.md to a sandbox path, shape-isomorphic to the personae under `api/server/personae/`. Personae differ from phase-agent skills: their YAML frontmatter contains an executable `decision_policy`…
原文の言語: 英語
Sub-skill of `compose-domain` (v3). Writes ONE phase-agent SKILL.md to a sandbox path, shape-isomorphic to existing skills under `api/server/skills/`. Persona authoring moved to `author-persona` in v3. Never writes to `api/server/skills/` directly.
原文の言語: 英語
v4 sub-skill #5 (optional). Adds the `ambient:` block — declares an AmbientAgent that watches the bus / graph / clock and spawns workflows of the listed types. Skipped when the operator answers "no" to "does this domain need an ambient hook?".
原文の言語: 英語
v4 sub-skill #2. Adds the `entities:` block to the brief and emits a per-domain projection function at `api/server/services/entity_projections/<workflow_type_snake>.py`. Validates entity kinds + relations against the Phase 1 `_NODE_TABLES` / `_REL_TABLES`…
原文の言語: 英語
Design-time meta-skill (v1). Given a persona brief (YAML) or a free-text idea, produce a complete Persona artefact set: a SKILL.md (frontmatter + executable decision_policy + prose body), a Python registry entry pasteable into `api/shared/personas.py`, and a…
原文の言語: 英語
Approves AP invoices and material expense claims within the controller band; escalates to CFO above £250k.
原文の言語: 英語