Pre-launch — Waitlist open

Raise the bar for your
entire workforce.

AI agents are becoming part of the workforce. But as capabilities spread across the organization, visibility disappears. SkillSpec changes that.

Join the Waitlist See How It Works
The problem

Teams are building skills and agents every day. Management has no way to assess them.

Across your organization, people are creating AI skills, prompts and agent instructions. Some are good. Some aren't. Many are duplicates. Nobody knows which perform best, whether any meet company standards, or whether changes actually improve outcomes.

Managers can't answer the most basic questions:

And that's just capability. Without assessment, you also can't know which skills carry hidden security risks. Skill poisoning — malicious instructions hidden in files referenced by a skill — has been demonstrated. Are your skills clean?

AI capability is managed
through guesswork.

The fix

SkillSpec is a competency management system for AI.

Assess, improve and govern the skills your agents use in production.

01

Assess

Measure skills against defined competency criteria — consistently, objectively, at scale.

02

Diagnose

Identify recurring weaknesses, capability gaps and performance trends across your skill portfolio.

03

Improve

Generate targeted improvements to the skill itself. Review and approve before anything changes.

04

Verify

Prove whether a change actually improved performance over time. No more hoping for the best.

SKILLSPEC — COMPETENCY DASHBOARD
Skills 24
Avg Score 78
Directives 3
Document Analysis
92 ↑ 4
Code Generation
81 ↑ 7
Email Drafting
74 →
Data Extraction
63 ↓ 5
Meeting Summary
58 ↓ 3
Image Description
34 ↓ 12
PENDING DIRECTIVES
⚡ IMPROVEMENT
Image Description scoring low on structured output. Proposed skill update targets format consistency.
30-DAY TREND
Platform

Built for teams managing AI at scale.

Visibility

A competency dashboard with proficiency tracking, trend analysis, and gap analysis across every agent and skill in your workspace.

Proficiency scores Trend charts Gap analysis Staleness signals

Automation

Agents can access their own skill context, review feedback, and apply approved improvements — closing the loop automatically.

Agent API Self-improvement CI/CD webhooks Quality gates

Collaboration

Build a workspace skill catalogue, starting with private drafts that can be shared for team-wide reuse.

Workspaces Private drafts Shared skills Skill catalogue

Security

Every skill gets a security advisory before hidden risks can take effect. Detect prompt injection, skill poisoning, and data exfiltration patterns.

Security advisory Poison detection Risk scoring Audit trail
Get early access

Replace guesswork
with a system.

Know which capabilities your organization possesses, which need improvement, and how they're evolving over time. Join the waitlist to get early access.