Loading...
Loading...
Found 484 Skills
Multi-step video annotation pipeline that turns raw videos into Chain-of-Thought training data — multi-level captions, structured descriptions, and QA pairs (MCQ, binary, open-ended) with reasoning traces, via VLM/LLM distillation. Use when the user wants to "create video training data", "generate video QA datasets", "build CoT reasoning traces from videos", "auto-label videos", or run the video_reasoning_annotation pipeline. Triggers include "video annotation", "video CoT", "video QA", "chain-of-thought", "video captioning pipeline", "video distillation".
Approve (submit an approving review) and squash-merge a PR given only its number, refusing when the QA gate or a blocking label forbids it. Routes fixable blockers to om-auto-fix-pr (red CI via its --ci-only mode, or conflicts and review problems via the full loop). Optionally file a follow-up issue at the same time. Use when the user says "approve and merge PR 123", "ship PR 123", or gives a PR number with intent to merge.
Grade a delivered claim against the artifact instead of against the report of the artifact. Use when acting as a verifier or reviewer on someone else's finished work, checking whether a fix actually landed, confirming a task's done result is true, auditing a claim that has no diff to read — a backfill, a migration, a deploy, a config change — or self-auditing your own claim before you publish it. Covers splitting a claim into checkable assertions, the three-state verdict (pass, fail, not-reached), and the specific failure modes that survive a careless check. Keywords verify, verification, grade, verdict, acceptance, prove it, did it actually work, confirm the fix, check the claim, QA sign-off.
Ultimate multi-agent framework for Google Antigravity. Orchestrates specialized domain agents (PM, Frontend, Backend, Mobile, QA, Debug) via Serena Memory.
Initializes a new webnovel project in deep mode, collecting full story/world/character/constraint data and generating all pre-writing files. Use when starting a new novel or running /webnovel-init.
Writes webnovel chapters (3000-5000 words). Use when the user asks to write a chapter or runs /webnovel-write. Runs context, drafting, review, polish, and data extraction.
Design security testing solutions, including OWASP Top 10, penetration testing, and vulnerability scanning. Default output is Markdown, and Excel/CSV/JSON output is available upon request. Use for security testing or security-testing.
Generates volume and chapter outlines from the master outline, inherits creative constraints, and produces writing-ready chapter plans. Use this when the user requests outlining or runs the /webnovel-plan command.
Utilize AI to assist in testing activities, including test data generation, defect root cause analysis, test prioritization, and intelligent test recommendation. The default output format is Markdown, and you can request Excel/CSV/JSON formats instead. This skill applies to AI-assisted testing scenarios.
Design API test plans and cases for REST/GraphQL/gRPC interfaces. Default output Markdown; can request Excel, CSV, or JSON. Use for API testing.
Automated changelog generation with semantic versioning. Triggers on: generating changelogs, updating versions, releasing software, version bumps, changelog updates, semantic versioning, and when users ask to "update changelog", "bump version", "create release", or "generate changelog". Use when user: needs to update CHANGELOG.md, bump package.json version, create a release, generate changelog entries, or follow semantic versioning and Keep a Changelog standards.
Recovers interrupted webnovel tasks with precise workflow state tracking. Detects interruption point and provides safe recovery options. Activates when user wants to resume or /webnovel-resume.