Skip to content

Product Roadmap

Last Updated: February 2026


Wedge → Expand → Dominate

Start with the most urgent pain (security), use it to attract developers (discovery), layer on value capture (monetization), and lock in enterprises (governance). Each phase builds on the previous one’s distribution.


Theme: Build the core infrastructure and scanning engine

#FeaturePriorityEffort
0.1SKILL.md parser (support full spec + all 26 platforms)P02 weeks
0.2Security scanner v1 (API key detection, known malware patterns, prompt injection checks)P03 weeks
0.3GitHub/ClawHub/SkillsMP ingestion pipelineP02 weeks
0.4Trust Score algorithm v1P01 week
0.5Basic web UI (search + scan results)P02 weeks
0.6CLI tool v1 (agentforge scan <path>)P01 week
0.7Developer accounts (GitHub OAuth)P11 week
0.8Landing page + docs siteP11 week
  • Build vs. Buy for scanning: Build our own static analyzer (proprietary moat) + integrate VirusTotal and Semgrep for known patterns
  • Database: PostgreSQL + pgvector for semantic search
  • Hosting: Start on managed cloud (AWS/GCP); plan for self-hosted enterprise later
  • Open source: Open-source the CLI scanner; keep the web platform proprietary
  • Can scan any SKILL.md file for security issues
  • Can ingest and index 100K+ skills from all major registries
  • Basic search works across all indexed skills

Theme: Become the trust layer for the agent skills ecosystem

#FeaturePriorityEffort
1.1Security Scanner v2 (deep analysis: dependency scanning, data flow analysis, permission mapping)P04 weeks
1.2Public Trust Score badges (embeddable — like “Snyk verified”)P01 week
1.3Scanning API (for platforms to integrate)P02 weeks
1.4Vulnerability disclosure and alerting systemP02 weeks
1.5Verified Publisher program (identity verification + code signing)P13 weeks
1.6”State of Agent Skills Security” report (Q2 2026)P12 weeks
1.7Browser extension (show trust score on ClawHub/SkillsMP/GitHub)P12 weeks
1.8CLI integration: agentforge install --verified-onlyP11 week
1.9Partnership outreach to ClawHub, SkillsMP, SmitheryP1Ongoing
1.10Security blog + thought leadership contentP2Ongoing
  • Publish “State of Agent Skills Security” report — drives PR and awareness
  • Open-source the CLI scanner — drives developer adoption
  • Offer free scanning to ClawHub and SkillsMP — partnership play
  • Target security-conscious developers and early enterprise adopters
MetricTarget
Skills scanned50,000+
CLI downloads10,000+
Scanning API integrations3+ platforms
Verified publishers500+
Monthly website visitors50,000+

Theme: Become the default place to find agent skills

#FeaturePriorityEffort
2.1Semantic search v2 (intent-based: “I need to automate email follow-ups”)P03 weeks
2.2Category taxonomy (20+ categories, human-curated)P02 weeks
2.3User reviews and ratings systemP03 weeks
2.4Curated collections (“Best DevOps skills”, “Top productivity skills”)P02 weeks
2.5Skill comparison view (side-by-side feature comparison)P12 weeks
2.6”Similar skills” recommendation engineP12 weeks
2.7Developer profiles and portfoliosP12 weeks
2.8Weekly “Trending Skills” newsletterP21 week
2.9API for agents to search Findable programmaticallyP12 weeks
2.10Findable MCP server (agents can discover skills through us)P02 weeks

Build an MCP server that any agent (Claude, OpenClaw, Codex) can connect to. When a user asks their agent to “find a skill for X,” the agent queries Findable directly. This makes us the discovery layer inside the agent itself — not just a website humans visit.

MetricTarget
Monthly active searchers100,000+
Skills indexed500,000+
Reviews submitted10,000+
Agent-initiated searches (via MCP)50,000/mo
Average session duration3+ minutes

Phase 3: Monetization Engine (Months 9-14)

Section titled “Phase 3: Monetization Engine (Months 9-14)”

Theme: Let developers make money from their skills

#FeaturePriorityEffort
3.1Stripe Connect integration (developer payouts)P03 weeks
3.2Pricing models: one-time, subscription, usage-based, freemiumP04 weeks
3.3In-agent purchasing (agent-initiated buy/subscribe flow)P04 weeks
3.4Developer revenue dashboardP03 weeks
3.5License key generation and managementP12 weeks
3.6UPI/India payments supportP12 weeks
3.7Promoted listings (self-serve ad platform)P13 weeks
3.8Affiliate/referral program for skill recommendationsP22 weeks
3.9Tax handling and invoicing (US, EU, India)P12 weeks
3.10Monthly payouts with detailed breakdownsP01 week
  • 15% for individual developers (lower than App Store’s 30%, competitive)
  • 20% for SaaS-connected skills (higher value, more infrastructure needed)
  • 10% for first year (promotional rate to bootstrap supply)
MetricTarget
Paid skills listed5,000+
Monthly GMV$500K+
Developers earning revenue1,000+
Average developer monthly payout$200+
Promoted listing revenue$50K/mo

Phase 4: Enterprise Governance (Months 14-20)

Section titled “Phase 4: Enterprise Governance (Months 14-20)”

Theme: Win enterprise budgets with governance and compliance

#FeaturePriorityEffort
4.1Private skill registries (org-only visibility)P04 weeks
4.2Policy engine (allowlists, blocklists, category restrictions, trust score thresholds)P04 weeks
4.3SSO integration (Okta, Azure AD, Google Workspace)P03 weeks
4.4SCIM provisioningP12 weeks
4.5Approval workflows (request → review → approve → install)P03 weeks
4.6Audit log (comprehensive: who installed what, when, data accessed)P03 weeks
4.7Compliance reports (SOC2, GDPR, DPDPA auto-generation)P14 weeks
4.8Cost management dashboardP12 weeks
4.9Admin console with role-based accessP03 weeks
4.10Self-hosted deployment optionP26 weeks
  • Target: Companies already using Claude Code, Codex, or OpenClaw at scale
  • Sales motion: PLG (start with free scanning) → team upgrade → enterprise contract
  • Initial verticals: Tech companies, financial services, consulting firms
MetricTarget
Enterprise customers100+
Average contract value$3,000/mo
Enterprise ARR$3.6M
Net revenue retention130%+
Self-hosted deployments10+

Phase 5: Agent Commerce Platform (Months 20-30)

Section titled “Phase 5: Agent Commerce Platform (Months 20-30)”

Theme: Power the transaction layer of the agentic economy

#FeaturePriorityEffort
5.1Agent-to-agent commerce protocol support (ACP, UCP, A2A)P06 weeks
5.2SaaS vendor skill optimization suite (Agent SEO)P04 weeks
5.3Transaction analytics for SaaS vendorsP13 weeks
5.4Agent recommendation API (agents ask “which CRM skill should I use?”)P04 weeks
5.5Bid-based ranking for commercial skillsP13 weeks
5.6Multi-agent skill orchestration (chain skills into workflows)P16 weeks
5.7Vertical skill catalogs (finance skills, healthcare skills, etc.)P24 weeks
5.8India market expansion (vernacular skill support)P24 weeks

By Phase 5, Findable is not just a marketplace — it’s the commerce infrastructure for the agentic economy. When AI agents need to select tools, services, or products to complete a task, they query Findable. SaaS vendors pay to be discovered and recommended. Enterprises govern what their agents can purchase. Developers earn revenue from their skills.


Month: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ... 30
├──────┤
Phase 0: Foundation
├───────────────┤
Phase 1: Security Wedge
├───────────────┤
Phase 2: Discovery
├──────────────────┤
Phase 3: Monetization
├───────────────────┤
Phase 4: Enterprise
├──────────┤
Phase 5: Commerce

PhaseEngineersDesignBiz/GTMTotal Team
Phase 03104
Phase 1411 (DevRel)6
Phase 25128
Phase 361310
Phase 4824 (incl. sales)14
Phase 5102618