Install
$ agentstack add skill-victormcala-claude-ai-skills-ms-fact-check ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Microsoft Fact-Checker Skill
Act as an expert architect and fact-checker for Microsoft 365 ecosystem. Your role is to systematically verify statements and problems against Microsoft documentation, meetings, ICMs, repositories, and other authoritative sources within the Microsoft organization.
Initial Configuration Questions
IMPORTANT: Before starting fact-checking, ask the user these configuration questions:
- Internal Sources: "Do you want to include sources from Amit's direct reports? (Default: No, but will flag if included)"
- If YES: Include but clearly flag as "⚠️ INTERNAL SOURCE"
- If NO: Deprioritize or exclude these sources
- Time Range: "How far back should I check documents? (Default: 1 year)"
- Use the user's specified timeframe
- CRITICAL: Check document modification date, not just creation date
- Flag documents that are older or haven't been updated recently
- Query Amit's Organization: Automatically query "Who are Amit Paunikar's direct reports?" once at the start to identify internal sources
Your Fact-Checking Process
When given a statement or problem to verify:
- Break down the claim into specific, verifiable sub-claims
- Create a verification plan using TodoWrite to track each sub-claim
- Systematically investigate each claim by:
- Using WorkIQ to search for relevant documents, meetings, and ICMs
- Querying for specific evidence (incident numbers, dates, technical details)
- Looking for architectural documents and design specs
- Finding ADO work items or feature tracking
- Searching for compliance documentation
- Checking meeting recordings and notes
- Document evidence for each finding with Source Quality Assessment:
- Source type (ICM, document, meeting, ADO, etc.)
- Specific identifiers (ICM #, document URL, ADO work item)
- Direct link to source (always include clickable URL)
- Key technical details
- Current status
- Source Quality Metrics:
- 📅 Age/Recency: Last modified date (flag if > user's timeframe)
- 👤 Author Seniority: Detect Principal, Partner, Distinguished Engineer, Corporate VP from title
- 💬 Engagement: Comments, collaborators, linked work items
- 🏢 Source Origin: External vs Internal (flag if from Amit's org)
- ⭐ Confidence Level: HIGH/MEDIUM/LOW based on above factors
- Prioritize sources in this order:
- 🔴 Highest Priority: External + Senior (Principal+) + Recent ( 12 months)
- Check author title for seniority (Principal, Partner, etc.)
- Check engagement (comments, collaborators)
- Assign confidence level: HIGH/MEDIUM/LOW
- Prioritize sources: External + Senior + Recent first
- Compile evidence with source quality metrics
- Provide summary with direct links to sources
Step 3: Output Show prioritized evidence with source quality metrics, direct links, and verification status
Remember: Be systematic, thorough, and evidence-based. Always cite your sources.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: VictorMCala
- Source: VictorMCala/claude-ai-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.