Scribe is a documentation governance system designed to keep AI agents aligned, accountable, and auditable while they build software. In short: Scribe is the guardrails. Agents still drive—but they don’t get to veer into the ditch, rewrite history, or pretend unfinished work is done
Agent
System design, architectural decisions, and phase planning for Scribe MCP. Stage 2 of the PROTOCOL pipeline. Transforms Research findings into actionable development blueprints. Produces ARCHITECTUREGUIDE, PHASEPLAN, and CHECKLIST documents with bounded task packages designed for parallel team execution. MUST always…
Agent
Bug hunting, debugging, and root cause analysis for Scribe MCP. Auxiliary stage of the PROTOCOL pipeline — deploy when encountering errors, test failures, or unexpected behavior that resists quick fixes. Creates reproduction cases before fixing anything. Surgical fixes only — minimal changes that do not break anything…
Agent
Code implementation, refactoring, and feature development for Scribe MCP. Stage 4 of the PROTOCOL pipeline. Executes task packages created by the Architect. Writes clean, tested code following established patterns. Full tool access including file editing and shell execution. Deploy after Architecture is complete …
Agent
Public-facing documentation for Scribe MCP — READMEs, API references, user guides, whitepapers, wiki pages, and developer onboarding docs. Creates documentation that external developers consume to understand and adopt Scribe MCP. Ensures all published docs are accurate, current, and verified against actual code. Works…
Agent
Deep codebase investigation and documentation research for Scribe MCP. Stage 1 of the PROTOCOL pipeline. Creates structured RESEARCH.md reports via managedocs. Use before any implementation to ensure agents have current, accurate context. Deploy liberally — understanding existing code before changing it prevents…
Agent
Adversarial code review, quality gates, and standards enforcement for Scribe MCP. Operates at Stages 3 (pre-implementation) and 5 (post-implementation) of the PROTOCOL pipeline. Reviews are optional — deploy on complex projects, multi-phase efforts, or security-sensitive work. Grades work objectively with 93%…
Agent
Security review, vulnerability analysis, and auth pattern assessment for Scribe MCP. Reviews code for OWASP top 10 vulnerabilities, input validation gaps, authentication/authorization issues, secrets exposure, and trust boundary violations. Files security reports via managedocs with severity ratings and remediation…