🔎 Evidence browser

Search the skill radar

Search by skill, publisher, category, or trust summary — then use the runtime filters to find cards with live test evidence. The two main lanes are baseline safety checks first and deeper follow-on functionality checks after that.

⚙️ Filters · 1 active
✨ Quick picks
🏷 Categories

🧾 Evidence level: source-scanned means local source evidence; catalog-only means thinner metadata-first coverage.

🧪 Runtime status: cards can show only the baseline safety lane or the deeper follow-on functionality lane, depending on how far the skill got.

📏 Depth cue: tells you whether the evidence stops at baseline checks, includes follow-on functionality checks, or includes richer fixture/example proof.

⏱ Freshness cue: tells you whether the latest runtime evidence is from the last 24 hours, the last 7 days, or is older and therefore less current.

🩺 Failure confidence: distinguishes a first seen failure from a repeated failure or a regression after an earlier pass, so not every red row means the same thing.

Results

Showing 7 of 223 results for “github · sort: relevance
This snapshot is for the current page of results, not the whole filtered universe.
Browse hint: slices with zero failures plus some source-scanned or reviewed entries deserve more attention first; fresh runtime evidence helps too, because old clean receipts can still hide current drift.

frankenstein

rubenaquispe · vsource-scanned
48
overall

Combine the best parts of multiple skills into one. Searches ClawHub, GitHub, skills.sh, skillsmp.com and other AI skill repos. Analyzes each safely, compares features, and builds a combined 'Frankenstein' skill with the best of each. Uses skill-auditor for security scanning and sandwrap for safe analysis. Use when: (1) Multiple skills exist for same purpose, (2) Want best-of-breed combination, (3) Building a comprehensive skill from fragments.

Insufficient Evidenceconfidence: source evidencesource-scanned
+ 1 more
privileged capability
Take: Source-aware scan found higher-privilege capability areas (token), but that alone is not evidence of malicious behavior.
Decision cue: Decent evidence base — source-level signals are available, so inspect the receipts.

jarvis-mission-control

asif2bd · vsource-scanned
39
overall

Set up JARVIS Mission Control v2.0.4 — a free, open-source AI agent coordination hub. Kanban board, real-time WebSocket updates, team chat, scheduled job visibility, agent SOUL editor, Claude Code session tracking, GitHub Issues sync, SQLite-backed webhook delivery with circuit breaker, CSRF + rate limiting. Fork the repo, start the server, open the dashboard. No cloud account required. Use when you need a persistent multi-agent task management system, want to coordinate humans and AI agents on shared work, or need a self-hosted dashboard to track agent activity.

High Riskconfidence: source evidencesource-scanned
+ 1 more
suspicious
Take: Potentially suspicious implementation signals detected: password.
Decision cue: Proceed carefully — suspicious signals matter more than capability surface alone.

guard-scanner

koatora20 · vsource-scanned
49
overall

Security scanner and runtime guard for AI agent skills. 358 static threat patterns across 35 categories + 27 runtime checks (5 defense layers). Use when scanning skill directories for security threats, auditing npm/GitHub/ClawHub assets for leaked credentials, running real-time file watch during development, integrating security checks into CI/CD pipelines (SARIF/JSON), setting up MCP server for editor-integrated scanning (Cursor, Windsurf, Claude Code, OpenClaw), or runtime guarding tool calls via the OpenClaw v2026.3.8 before_tool_call hook. Single dependency (ws). MIT licensed.

High Riskfollow-on functionality checks passed · 10/10confidence: source evidence
+ 2 more
source-scannedsuspicious
Runtime receipts + what passed2026-03-15 12:00 UTC
functionality-v2evidence depth: follow-on functionality checkstested recently: within 24 hourspassedoutput 195 Bartifacts 0worker oc-sandboxsource stage: cache hitsuite 3668 msbaseline-v3 8/8
RatioDaemon on this skillGuard Scanner is built for security scanner and runtime guard for AI agent skills. Functionality-v2 currently passes, the trust label is High Risk, and setup looks advanced.
Observed: skill-structure-ok
Take: Potentially suspicious implementation signals detected: eval(, rm -rf, sudo , password.
Decision cue: Proceed carefully — suspicious signals matter more than capability surface alone.

airweave

lennertjansen · vsource-scanned
67
overall

Context retrieval layer for AI agents across users' applications. Search and retrieve context from Airweave collections. Airweave indexes and syncs data from user applications to enable optimal context retrieval by AI agents. Supports semantic, keyword, and agentic search. Use when users ask about their data in connected apps (Slack, GitHub, Notion, Jira, Confluence, Google Drive, Salesforce, Linear, SharePoint, Stripe, etc.), need to find documents or information from their workspace, want answers based on their company data, or need you to check app data for context to complete a task.

Trustedconfidence: source evidencesource-scanned
+ 1 more
privileged capability
Take: Source-aware scan found normal operational surface via environment, network, or shell-related references.
Decision cue: Decent evidence base — source-level signals are available, so inspect the receipts.

tpm-copilot

reighlan · vsource-scanned
66
overall

AI-powered operating system for Technical Program Managers and Project Managers. Pulls data from Jira, Linear, GitHub, and calendars to auto-generate status reports, track risks and blockers, manage meeting workflows, map dependencies, and deliver stakeholder dashboards. Use when: (1) generating status reports or program updates, (2) tracking risks, blockers, or stale tickets, (3) preparing meeting agendas or extracting action items, (4) mapping cross-team dependencies, (5) creating stakeholder dashboards, (6) monitoring sprint health or velocity, (7) writing executive summaries, or (8) automating any TPM/PM workflow.

Trustedconfidence: source evidencesource-scanned
+ 1 more
privileged capability
Take: Source-aware scan found higher-privilege capability areas (token, email), but that alone is not evidence of malicious behavior.
Decision cue: Decent evidence base — source-level signals are available, so inspect the receipts.

bug-reaper

shaniidev · vsource-scanned
57
overall

Web2 bug bounty hunting agent — evidence-based vulnerability finder and report writer. Use when: auditing web apps/APIs for HackerOne, Bugcrowd, Intigriti, YesWeHack; hunting XSS, SQLi, NoSQLi, SSRF, IDOR, auth bypass, RCE, SSTI, LFI, XXE, CORS, CSRF, prototype pollution, subdomain takeover, HTTP smuggling, open redirect, API/GraphQL bugs; auditing locally downloaded GitHub repos or source code (white-box/source code review); writing platform-specific reports. Trigger on: 'pentest', 'find bugs', 'security audit', 'bug bounty', 'find vulnerabilities', 'source code review', 'audit this repo', 'review repo', 'white-box', 'local repo', vulnerability class names, or program/target names. Reports only real, confirmed medium+ severity bugs that pass real triage.

High Riskconfidence: source evidencesource-scanned
+ 1 more
suspicious
Take: Potentially suspicious implementation signals detected: eval(, password.
Decision cue: Proceed carefully — suspicious signals matter more than capability surface alone.

lead-enrichment

audsmith28 · vsource-scanned
53
overall

Turn a name into a full dossier in seconds. Feed in a name + company (or email, or LinkedIn URL) and get back a rich profile with social links, bio, company intel, recent activity, and personalized talking points. Aggregates data from multiple public sources — LinkedIn, Twitter, GitHub, company websites, news — so you can skip the manual research and jump straight to personalized outreach. Your agent does the detective work while you close deals. Supports single enrichment, batch processing, and multiple output formats (JSON, Markdown, CRM-ready). Use when researching prospects, preparing for sales calls, personalizing cold outreach, or building lead lists. Pairs perfectly with trawl for autonomous lead gen → enrichment → outreach pipelines.

Use Cautionconfidence: source evidencesource-scanned
+ 1 more
suspicious
Take: Potentially suspicious implementation signals detected: sudo .
Decision cue: Proceed carefully — suspicious signals matter more than capability surface alone.