Mint tamper-evident, post-quantum-signed receipts for consequential agent actions and verify them offline from the certificate alone.
Read-only audit of the skills, MCP servers, hooks, plugins, and credentials already installed across Claude Code, Codex, OpenClaw, and Cursor, with a structured risk briefing.
Installs and refines a repo quality framework where every agent rule names a gate that fails a commit, push, or CI run.
Models the true margin and TCO of each go-to-market channel and designs partner tiers, rebates, and channel-conflict rules.
Drives the Compass CLI's local code knowledge graph so Claude answers architecture, dependency, and change-impact questions with cited evidence.
Compresses an approved game design and art direction into a minimal BUILD_BRIEF and produces a runnable game candidate on the target runtime.
Decides which model, effort level, and cascade shape each subagent gets, routing on measured cost per completed task instead of per-token price.
An expert skill for EU CSRD/ESRS work: scope determination, double materiality assessment, gap analysis and disclosure drafting with article-level citations.
Turns Claude into a CMMC 2.0 / NIST SP 800-171 consultant for gap assessments, SSP drafting, SPRS scoring and POA&M management in the US defense supply chain.
Drive a kitty-graphics terminal app headlessly in a pty and produce a recorded video, timeline and checks summary as the deliverable.
A master entry point that chains eight sub-skills — topic, writing, review, formatting, images, and publishing — to take a WeChat Official Account post from idea to draft box in one request.
Scans AI coding-agent caches (.claude, .cursor, .antigravity, .openclaw) or any folder you name, then turns scattered experiment logs into PaperOrchestra-ready idea.md and experimental_log.md.
An iterative LaTeX paper refinement loop that simulates peer review and only keeps revisions that provably raise the score.
Turns your idea notes, experiment logs, LaTeX template and conference guidelines into a strict JSON paper outline with figure, literature-search and section plans.
A meta-skill that triages skill requests, forces a failing baseline before creation, and ships every new skill with regression evals.
A design-first workflow skill that sketches types, signatures, and module boundaries before coding, then keeps the sketch honest during implementation.
Fans out N parallel agents on the same task, cross-judges them against a rubric, picks the strongest base, and grafts the losers' best ideas into one verified artifact.
Plays a French DGFIP tax inspector, reviewing a company's books across nine audit axes and drafting a formal reassessment report.
Runs a 7-phase, NEP-compliant statutory audit of a French company's annual accounts and issues a structured audit report with an opinion.
Final-stage assembly skill that lays narration audio over the source video, ducks the original sound, generates and burns subtitles, and normalizes loudness into a finished recap.