Make AI agent reasoning visible and auditable as an interactive graph. Arcgram turns walls of prose into typed nodes and explicit edges you inspect and correct node by node. Use it as a visual agent harness: propose → draw → self-audit → fix, human in the loop. Outputs one standalone HTML file: no build step, no npm…
★not rated 13
changed 24d agoA
tokens not measured
originalApache-2.0
Optimizes and rewrites prompts using proven prompt engineering techniques: draft diagnosis, canonical structure, technique selection by task type and target model (Claude 4.x, OpenAI, Gemini), with worked examples, a quality checklist, and an evaluation guide.
★not rated 13 5mo agoA
tokens not measured
originalMIT
Search the PubChem chemical database for compounds, properties, safety data, bioactivity, cross-references, and entity summaries via MCP. STDIO or Streamable HTTP.
AI-First Product Manager skills: synthetic personas, persona interviews (exploration and validation), insight extraction, persona critique panels, and exposure plans.
★not rated 13
changed 13d agoA
tokens not measured
Publish single-page HTML to the public web instantly via ht-ml.app — create, update, upload assets, optional password protection, and 20 ready-made page templates.
Answers Kotlin/Java library questions from the real published sources jar — resolved signatures, KDoc, dependency tree and source search — instead of recalling them.
★not rated 13
changed 12d agoA
tokens not measured
originalApache-2.0
Enterprise-grade AI operating system for hotel teams — revenue management, forecasting, P&L analysis, labor optimization, reputation intelligence, competitive intel, sales pipeline, content generation, SOP creation, and owner reporting. Built by HospitalityOS for GMs, revenue managers, finance directors, and ownership.
A collection of read-only DevOps advisor skills. Point them at infrastructure, clusters, pipelines, databases, or an incident and they investigate, base findings on evidence, and write structured remediation or investigation plans any agent can execute. They never apply changes themselves.
A plugin for Yonsei University electronic attendance records, used with the “오늘의 연세·포털 연결” add-on. It reads attendance information and can summarize attached screenshots or tables.
Engineering decision framework based on Elon Musk's 5-step algorithm: question requirements, delete parts, simplify and optimize, accelerate cycle time, automate. Decomposed into 6 skills with strict order enforcement and Obsidian-compatible wiki links.
★not rated 13 5mo agoA
tokens not measured
originalMIT
AI video pipelines for Claude Code: a live-action promo-reel editor (demucs/whisper/rembg + ffmpeg compositing), a code-driven motion-graphics film renderer (GSAP + headless Chromium + ffmpeg), plus campaign mode — music-first beat-locked scripting, keyless CC asset casting, and multi-agent parallel reel builds. Drive.
Understand any codebase, then ship changes to it. Read-only comprehension tools (maprepo, blastradius, explainsymbol, investigate, localize, regressiongaps, rootcause, docsfor) hand you engineering decisions — what breaks if you change X, what's untested, where a ticket or bug lands, which docs describe it — from.
★not rated 13
changed 7d agoA
tokens not measured
originalMIT
Requirements-driven development workflow in the spirit of DO-330 TQL-5. Formal Gherkin-style TOR requirements (TOR-NN-XXXXXXX) serve as the single source of truth — epics implement requirements, tests are derived from Given/When/Then, and wrapup verifies against the requirements baseline rather than prose acceptance…
Work with turbopuffer, a serverless vector and full-text search database. Includes guidance for querying (vector ANN, BM25, hybrid), writing, schema management, native embeddings, sharding, pinning, branching, bulk ingestion, and troubleshooting.
★not rated 13
changed 1mo agoA
tokens not measured
Audit any website across SEO, answer-engine and generative-engine criteria, and generate llms.txt, llms-full.txt and facts.json so AI engines can ingest it as ground truth.
★not rated 13▲
+1 2mo agoA
tokens not measured
originalMIT
Skills covering the U.S. EB-2 National Interest Waiver from first question to filed response. Decide whether to pursue it and under what proposed endeavor (niw-evaluate); plan the evidence to gather (niw-evidence-plan); find government sources showing the endeavor advances a national priority (niw-evidence-finder)…
★not rated 13
changed 2mo agoA
tokens not measured
Defensive IOC + vulnerability triage on free/open-source intel: a subagent plus two skills — threat-intel-lookup (aggregates 10 key-free feeds and looks up IPs, domains, URLs, CVEs) and exploit-availability-check (rates a CVE's exploit/PoC maturity via KEV, EPSS, Metasploit, Nuclei, Exploit-DB, and PoC aggregators).…
Agent-native FP&A toolbelt for learning a company, building its data access and models, preserving memory, and improving forecasts against actual outcomes.
★not rated 13▲
+2 4mo agoA
tokens not measured
originalMIT
Architecture Decision Record toolkit for AI coding agents: bootstrap, multi-format authoring, bounded deterministic enforcement, lifecycle workflows, context injection, generated graph indexes, and local health checks.
★not rated 13
changed 18d agoA
tokens not measured
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: