Loading...
Loading...
Found 12 Skills
Add a Jev-powered decision and release gate after an office assistant generates a report, proposal, spreadsheet, document, or slide deck. Use for claim-to-evidence checks, narrow requirement decisions, risk triage, and human-review routing; do not use it to generate the deliverable or claim that Jev browsed or verified a URL.
Optional skill. Reconstruct a human-review-preparation file from an existing pull request, merge request, branch diff, or commit range in a repository the user trusts. Use when the user wants retrospective understanding of already-implemented changes, AI-side assessment and recommendations, and an optional provider-specific sharing variant written to a local file when needed.
Use Verify for snapshot testing in .NET. Approve API surfaces, HTTP responses, rendered emails, and serialized outputs. Detect unintended changes through human-reviewed baseline files.
TranscribeMe platform help — human+AI hybrid transcription service with 99% accuracy guarantee on human-reviewed tiers, HIPAA compliance, and REST API. Use when uploading audio for human-edited transcription, choosing between Machine Express and human-reviewed tiers, integrating TranscribeMe API into a transcript pipeline, comparing TranscribeMe vs Rev vs GoTranscript vs Sonix for accuracy-critical transcription, troubleshooting slow turnaround or inconsistent punctuation, setting up HIPAA-compliant medical or legal transcription, or debugging chunked upload or OAuth token issues. Do NOT use for live meeting recording or real-time transcription (TranscribeMe is upload-only — use /sales-note-taker for live meeting tools).
The OpusClip craft skill — turn long video (podcasts, webinars, interviews, streams) into publishable short clips with AI candidate-finding, honest triage, and a human-reviewed pipeline. Use when someone wants to use OpusClip, turn a podcast/webinar/VOD into shorts, understand the Virality Score, pick a plan, pull a moment from a long recording (ClipAnything moment search), or automate a clipping pipeline. Uses the CLIPS framework. Reads captions-and-clipping + the source skill + brand-profile first. The agent plans the pipeline, credit math, and QA; the HUMAN reviews every clip and approves; WoopSocial publishes the finished exports (it does not clip or edit video). Integrity spine: the Virality Score is a PREDICTION for triage, never truth; 1 credit = 1 SOURCE minute; no unreviewed auto-post pipelines; no out-of-context clips. Distinct from captions-and-clipping (general craft), descript (long-form edit), capcut (short-form polish), and the platform publishing skills.
Create and composite HeyGen digital-human videos in the content workspace. Use whenever the user asks for a 数字人, 数字人5, HeyGen, AI 口播人物, 数字人小窗, 圆形头像口播, full-screen digital human, or asks to make a video that uses the existing avatar. Enforce Creator-plan CLI OAuth, the configured Digital Twin, local IndexTTS2 narration, Avatar III for an unspecified digital human but Avatar V whenever the named preset 数字人5 is requested, adaptive face measurement plus the approved human-reviewed visual-centering offset in the headroom_08 circular bottom-left layout unless full-screen is explicitly requested, subtitle-safe placement, credit-aware master reuse, final QC/archive routing, and mandatory pairing with `heygen-avatar-motion` before any motion prompt or paid generation.
Spec-driven feature work that freezes obligations instead of the plan: one human-reviewed plan with EARS criteria, path, entities, interface and one-way doors, then proof-backed checks, then build, then an independent Verifier. Use when the user says "tlc-spec-lean", "plan feature", "specify feature", "write the checks", "build this plan", or "verify work". Do NOT use for standalone design documents unattached to a feature, architecture decomposition analysis, or work that already has a task list or checklist to execute.
Launch the mathematical modeling competition workflow, confirm competition and typesetting preferences, and initialize the plan, to-do list, and stage artifacts. When CUMCM 2026 is identified, enable human decision gates, AI usage traceability, official contest search restrictions, and the submission packaging process.
Execute an approved implementation plan phase by phase with verification and human gates.
Controlled plan execution with human review checkpoints - loads plan, executes in batches, pauses for feedback. Supports one-go (autonomous) or batch modes.
Analyze mathematical modeling competition problems and attachments, clarify ambiguities, design and compare candidate models, and develop an implementable solution through quick experiments and independent review. In CUMCM 2026 mode, enforce human approval gates for problem interpretation and core modeling while recording AI usage.
Deduplicate CRM accounts and contacts and keep them duplicate-free: audit company and person identity, run recurring deduplication plays directly on CRM models, merge safe exact matches, and route uncertain clusters to manual review. Triggers: "deduplicate our CRM accounts", "deduplicate CRM contacts", "our CRM has duplicate people", "merge duplicate contacts in HubSpot", "we keep creating duplicate account records", "merge duplicate companies in HubSpot", "set up recurring CRM deduplication", "review ambiguous duplicates". HubSpot, Salesforce, Attio, Slack, LinkedIn profiles, phone numbers, Cargo project, findRecords, Scoring, Human Review, mergeRecords. Skip when: the request is to add or refresh CRM data rather than merge duplicate records; use crm-enrichment.