Forge Project Cleanup Charter Alignment
Charter alignment matrix for the active project workflow, per Spec §9. This matrix updates the pre-cleanup assessment from the Research document to reflect the post-Plans 1-3 state.
Alignment Matrix
| Charter principle / constraint | Compliance | Active evidence (post-cleanup) | Gap | Correction |
|---|---|---|---|---|
| Forge is the single source of truth for organizational vision, plans, procedures, and agents. | ✅ Compliant | Project work-products and Forge knowledge docs are durable markdown; Bot configuration and prompt versions are centrally managed; todos tool group replaces flat-file queues; WIP removed; bot-prompt compliance audit confirms alignment; prompt-version artifacts provide inspectability | None | Plans 1-3: Removed flat-file Erik_Todo/Agent_Todo/WIP as active state stores; migrated to todos tool group; added bot-prompt compliance checks |
| Uvilo-specific information lives in department folders that plug into the Forge framework. | ✅ Compliant | Department/project hierarchy is used consistently: Forge/Projects/{Project} and Forge/Skills/*; project naming convention {Department}/{ProjectShortName} enforced | None | No correction needed — was already compliant |
| Human sovereignty: humans make decisions that matter; agents execute. | ✅ Compliant | User-configurable human gates (Spec §12) with documented default set (Spec_Eval, Plan_Eval, Verify); Runner enforces gates; auto-dispositions logged as reversible informational items; todos tool routes human decisions with structured body content and artifact links | None | Plan 3: Updated project-runner prompt with full human gate options table, default gates, frontmatter storage, and enforcement rules |
| Clarity: decisions are traceable from research through specification to outcome. | ✅ Compliant | Lifecycle has Vision → Research → Spec → Plan → Eval reports; evaluations are persisted; work-product communication rules (Spec §4) enforce that substantive content lives in durable artifacts, not conversations; eval criteria include work-product communication check | None | Plans 1-3: Removed WIP and conversation duplication; added work-product communication rules to all bot prompts; added eval criterion for work-product communication |
| Stewardship: consider customers, personnel, management, shareholders, and community. | ✅ Compliant | Project Vision and Spec can include stakeholder requirements; Charter alignment matrix explicitly reviews stewardship; Vision_Eval checklist item 5 requires evaluator to check if stewardship applies and flag missing requirements | None | Added charter considerations to Vision_Eval checklist and Vision template — if a project affects stakeholders, the evaluator flags it and a requirement is added |
| Continuous improvement: processes research, implement, and improve themselves on an ongoing cycle. | ✅ Compliant | Project lifecycle includes Extract, Extract_Eval, Forge Optimizer guidance, persisted reports, and this cleanup project itself; Inngest reconciliation provides safety net | None | No correction needed — was already compliant |
| Accessibility: power without complexity. | ✅ Compliant | User Guide updated; Project Runner abstracts phase dispatch; todos tool group provides clean action interface; WIP and flat-file todo details removed from user-facing docs; human gates are user-configurable with sensible defaults | None | Plans 1-2: Removed stale WIP and Erik_Todo implementation details from user-facing docs; Plan 3: Added user-configurable human gates |
| Goal: serve as the single source of truth for all organizational information. | ✅ Compliant | Durable markdown docs and project work-products support this; todos tool group is the canonical queue; WIP removed; bot prompts aligned with Knowledge; Phase file is the canonical state signal | None | Plans 1-3: Consolidated state into canonical sources; removed stale state stores |
| Goal: structure the organization into departments with cross-functional links. | ✅ Compliant | Department roots, Projects/, Skills/, and archived project structure are documented; project naming convention includes department | None | No correction needed — was already compliant |
| Goal: automate research, implementation, and continuous improvement of organizational processes. | ✅ Compliant | Project Thinker/Worker/Evaluator/Runner roles automate Research, Plan, Execute, Eval, Verify, Extract, and Complete; Inngest provides scheduled reconciliation | None | No correction needed — was already compliant |
| Goal: run as many organizational functions as possible. | ✅ Compliant | Project system runs project management and implementation workflows through agents | None | No correction needed — was already compliant |
| Goal: provide an easy, repeatable process to plan, create, and manage initiatives. | ✅ Compliant | Standard lifecycle and Project Flow skill provide repeatability; documentation is consistent across Workflow, User Guide, Skills, and bot prompts; WIP and flat-file todo details removed | None | Plans 1-2: Normalized lifecycle wording; removed stale WIP/todo references |
| Goal: optimize organizational performance and success. | ✅ Compliant | Forge Optimizer and run summaries exist; context boundaries reduce waste; run/cost evidence is recorded when available; Vision_Eval checklist item 5 requires evaluator to check if performance tracking applies and flag missing requirements | None | Added charter considerations to Vision_Eval checklist — if a project involves automation or infrastructure where cost/token metadata matters, the evaluator flags it and a requirement is added |
| Scope: Forge is an orchestration and partial execution layer, not a replacement for specialized external systems. | ✅ Compliant | Orchestration doc describes thin dispatcher; Worker uses external tools; Forge verifies outcomes | None | No correction needed — was already compliant |
| Scope: integrate mature external tools rather than rebuild them. | ✅ Compliant | spawnAgent, MCP tools, Typesense, Postgres, Inngest, and Bot API integration are documented; todos tool group replaces flat-file implementation; no flat-file todo/checklist remains as active workflow | None | Plans 1-3: Replaced flat-file todo/checklist with todos tool group |
| Scope: Forge handles specification, planning, and verification; specialized execution is delegated and checked. | ✅ Compliant | Workflow separates Spec, Plan, Execute, Execute_Eval, Verify; Evaluator checks Worker outcomes | None | No correction needed — was already compliant |
| Scope: Forge must publish outward according to organization-level visibility. | ✅ Compliant | Astro/Starlight docs, sidebar convention, and frontmatter visibility exist; Page Manager handles page/sidebar changes; sidebar auto-generation from file structure | None | No correction needed — was already compliant |
| Scope: designed for smaller AI-first organizations where human/agent boundaries evolve. | ✅ Compliant | Interactive vs autonomous Runner modes; user-configurable human gates; auto-fix loops support changing human involvement; WIP removed (conversations persist, making WIP obsolete) | None | Plans 1-2: Removed WIP as active workflow state; Plan 3: Added user-configurable human gates |
| Requirement: provider independence; components must be replaceable through interfaces. | ✅ Compliant | Bot API, spawnAgent, and MCP-style tool discovery create interface boundaries; todos tool group provides stable interface for human/agent queues | None | Plans 1-3: Removed LibreChat-derived flat-file todo references; todos tool group provides provider-independent interface |
| Requirement: cost efficiency; token and infrastructure costs measured and optimized; agents carry only needed context. | ✅ Compliant | Project Flow and phase skills define context boundaries; Agent Architecture emphasizes on-demand skills and tool loading; work-product communication rules enforce concise handoffs; run/cost evidence field in handoff packet | Run/cost metadata unavailable from platform (noted explicitly) | Plans 1-3: Enforced concise handoff payloads with artifact links; removed WIP; added run/cost evidence to handoff packet; explicitly note when exact data is unavailable |
| Requirement: tool integration; connect with organization platforms/tools. | ✅ Compliant | todos tool group replaces flat-file Erik_Todo.md/Agent_Todo.md; all active instructional references to flat-file todos removed from Knowledge, Skills, Guides, and bot prompts | None | Plans 1-3: Replaced flat-file todo workflows with todos tool group; removed all active instructional references |
| Requirement: multi-user with automatic merging, version control, and archival. | ✅ Compliant | Git-backed markdown supports version control; archived projects are documented; todos tool group provides user-scoped tasks; append-safe historical work-products documented | None | Plans 1-3: Migrated from flat-file per-user todos to todos tool group with proper user scoping |
| Requirement: accountability; every project has an owner and approver. | ✅ Compliant | Frontmatter includes owner/approver; project docs use owner: "erik@uvilo.com"; todo items include responsible owner; handoff packets include responsible bot/agent role | Templates allow blank approver (explicitly pending) | Acceptable — approver can be pending until human gate is reached. Todo items and reports include responsible owner. |
| Requirement: transparency; every decision traceable; Research shows conclusions; Specs show final decisions; both preserved. | ✅ Compliant | Research, Spec, eval reports, and historical append-only semantics support traceability; work-product communication rules prevent conversation-only decisions; prompt-version artifacts provide bot prompt traceability | None | Plans 1-3: Added work-product communication rules; added eval criterion for work-product communication; created prompt-version artifacts |
| Requirement: resilience; recover from agent errors/tool failures; state persists; work never lost to crashed conversation. | ✅ Compliant | Phase file, reports, git, persisted conversations, and AgentJob monitoring provide resilience; recovery contract documented (Phase + Execute_State + durable work-products + persisted conversation/job metadata + git status + todo items); WIP removed as redundant state store | None | Plans 1-2: Removed WIP from active flow; documented recovery contract; Plan 3: Updated bot prompts with explicit recovery instructions |
| Requirement: evolvability; meta-process for improving Forge itself. | ✅ Compliant | Project lifecycle includes Extract and Forge Optimizer; this cleanup project is an explicit meta-process; Charter alignment matrix provides ongoing compliance check | None | No correction needed — was already compliant |
| Constraint: security is non-negotiable. | ✅ Compliant | Agent Architecture documents bot access control and least-privilege tools; Vision_Eval checklist item 5 requires evaluator to check if security applies and flag missing requirements | None | Added charter considerations to Vision_Eval checklist — if a project touches sensitive data or privileged tooling, the evaluator flags it and a requirement is added |
| Constraint: privacy serves the mission and is an organizational decision. | ✅ Compliant | User/conversation context has opt-in write policy; bot access is scoped; Vision_Eval checklist item 5 requires evaluator to check if privacy applies and flag missing requirements | None | Added charter considerations to Vision_Eval checklist — if a project handles user data or personal information, the evaluator flags it and a requirement is added |
| Constraint: simplicity of use, not of system. | ✅ Compliant | Project Runner hides orchestration; User Guide gives human instructions; todos tool provides clean action interface; WIP and flat-file todo details removed from user-facing docs; human gates have sensible defaults | None | Plans 1-3: Moved user-facing gates to todos system; removed stale implementation details |
| Constraint: inspectability by humans; personnel can view skills and agent internals. | ✅ Compliant | Skills and Knowledge are markdown; Bot prompt versions can be reviewed via prompt-version artifacts; bot-prompt inventory document exists; infrastructure inventory provides complete component listing | None | Plan 3: Created prompt-version artifacts; maintained bot-prompt inventory |
Summary
| Status | Count |
|---|---|
| ✅ Compliant | 29 |
| ⚠️ Partial | 0 |
| ❌ Violation | 0 |
The cleaned workflow has no known unaddressed Charter violations. ✅