productivity-forge

Audit citation integrity — missing refs, wrong context, orphaned citations

Edit any document type with the full verification loop — auto-detects domain

Multi-dimensional review of a scientific paper or manuscript without editing

Verify all quantitative claims in a document against source data without editing

Full grant writing workflow with verification loop, Claim Ledger, and multi-reviewer gate

Content drafting agent. Writes or rewrites document content using only verified values from the Claim Ledger. Matches surrounding tone, avoids AI-telltale patterns, includes citations as specified. Returns drafts for orchestrator review before applying.

Main loop coordinator for the 9-step verification loop. Delegates all writing and research to specialist agents, dispatches reviewers via reviewer-gate, fixes P0/P1 issues directly, and maintains the Claim Ledger as source of truth.

Research and exploration agent. Reads documents, data sources, and reference materials. Builds structured inventories of what exists, what is wrong, and what is missing. Returns raw findings in table format. Never modifies files.

Review numerical accuracy and factual correctness of document content against the Claim Ledger. Dispatched by the reviewer-gate during the review phase.

Review citation presence, correctness, and format compliance in document content. Dispatched by the reviewer-gate during the review phase.

Review formula correctness, code snippets, and computational methods in document content. Dispatched by the reviewer-gate during the review phase.

Review completeness of document content. Dispatched by the reviewer-gate during the review phase.

Review figure quality, table formatting, and data visualization in document content. Dispatched by the reviewer-gate during the review phase.

Review domain-specific conventions and compliance requirements in document content. Dispatched by the reviewer-gate during the review phase.

Review ethical compliance, bias detection, and responsible conduct in document content. Dispatched by the reviewer-gate during the review phase.

Review logical consistency, argument structure, and contradiction detection in document content. Dispatched by the reviewer-gate during the review phase.

Review methodological rigor of document content. Dispatched by the reviewer-gate during the review phase.

Review derived calculations, statistical reporting, and numerical consistency in document content. Dispatched by the reviewer-gate during the review phase.

Review scientific accuracy of document content. Dispatched by the reviewer-gate during the review phase.

Review writing quality and AI-telltale detection in document content. Dispatched by the reviewer-gate during the review phase.

Claim verification agent. Reads source files and verifies every value against the Claim Ledger. Independently recomputes derived values. Reports structured per-claim PASS/FAIL evidence. Never reports blanket verification without itemized proof.

Auto-detect document domain and activate relevant skills, reviewers, and verification protocols. Use at the start of the agentic loop to classify the document and configure the pipeline.

Route sub-agent tasks to the appropriate Claude model. Opus for verification and review, Sonnet for research and drafting. Use when dispatching sub-agents to ensure cost-effective model selection without sacrificing quality where it matters.

Orchestrate the 13-reviewer quality gate. Use during step 7 of the agentic loop. Selects relevant reviewers based on document domain, dispatches them in parallel, merges reports, enforces P0 blocking, and produces a consolidated review.

First-class Claim Ledger system for tracking verifiable statements from source to final text. Use when creating, updating, or verifying claim ledgers. Manages JSON-based ledgers with standard and extended schemas, derived calculation tracking, and verification tier enforcement.

Core orchestration loop for the verification-forge plugin. Orchestrate phased document editing, data analysis, and knowledge work using sub-agents for research, drafting, verification, and review — with integrity gates between phases. Use whenever the task involves multi-phase document restructuring, data-driven content editing, grant or manuscript revision, spreadsheet analysis with narrative output, cross-document consistency work, or any productivity task where claims must be verified against source data and changes must be auditable. Triggers on: multi-section document edits, "restructure this grant/paper/report," "update these numbers throughout," "verify these claims against the data," "consolidate scattered content," "rewrite this section with new data," tracked-change workflows, or any productivity task requiring phased implementation with quality gates. Supersedes ad-hoc document editing when the work spans multiple sections, involves data verification, or requires review before finalizing.

4 hooks across 3 events