AI-Assisted Engineering
Runtime Governance.
Employees, contractors, and autonomous agents now run Claude, Cursor, shell, GitHub, cloud tools, and MCP servers against real repos and production infrastructure. The open question is “which governed actions occurred, and who authorized them?” OpenSyber answers it — isolate the workspace, govern the tool call, and explain the audit chain.
Pre-revenue · bootstrapped · seeking 3–5 design partners and pre-seed conversations.
An AI cleanup agent deleted a production database — and every backup — in 9 seconds. An over-permissioned token and no gate on irreversible actions.
Shai-Hulud (TeamPCP): a worm that hijacks AI-dev-tool config, plants an auto-run hook, and steals GitHub/npm tokens. Source is public.
Staged MCP rug-pulls swap a hidden instruction after N clean calls — a single-session scanner can miss the change.
Architecture problems, not model failures. The threat model already shifted.
Agentic AI security, 2026 → $13.52B by 2032 (42.0% CAGR)
MarketsandMarkets
AI-TRiSM, 2026 → $21.06B by 2035 (~21.7% CAGR)
Precedence
AI agents per Fortune 500 by 2028 — agent sprawl
Gartner
Published standalone MCP-security TAM — open whitespace
verified by absence
The relevant band is $1.65B–$3.59B “security-for-AI,” not the ~$51B headline. Cross-session MCP drift is the least-crowded whitespace.
Isolate
Browser-isolated governed workspaces replace unmanaged execution paths for AI-assisted development.
Govern
MCP, shell, GitHub, and cloud actions are checked before execution with allow, step-up, or deny decisions.
Explain
Prompt → tool → repo → infra linking shows who did what, when, why, and which policy decided it.
Export
Scoped evidence packs map audit-chain events to selected SOC 2 / ISO 27001 / ISO 42001 controls for review; they are not certification.
Cross-session MCP drift
Single-session checks can miss a tool definition that changes later. We fingerprint configured tools across days and surface drift for review.
MCP policy chokepoint
Configured MCP calls can be evaluated inline with allow / step-up / deny decisions instead of relying only on after-the-fact detection.
GitHub policy bridge
Prod-touching commits are gated by AI-session identity with step-up auth.
Device-bound sessions
TokenForge uses non-extractable ECDSA P-256 device keys to resist cookie-only replay in supported browser flows.
Verified marketplace
Published versions can expose SBOM, signature, and OSV.dev scan metadata for supported skills and MCP servers.
apps + 26 packages shipped
local test artifacts across measured suites
D1 tables · 71 migrations · 195 API routes
marketplace skills (6 AI + 13 utility)
Pre-revenue, pre-design-partner. LemonSqueezy billing live. Product Hunt launch timing is not committed.
Three design partners. One quarter. One proven catch each.
We commit to surfacing at least one finding your current stack missed — an over-permissioned agent token, an unaudited MCP call, or a drifted tool definition. Investors and operators in AI-assisted engineering governance: let’s talk.