Quality

Growth Reviewer checks Diffmode growth-tactics outputs against the matching rubric for a given dimension. Use it to gate enrichment, think-tank, or synthesis outputs before downstream stages consume them.

I18n Validate checks locale files for missing translations, orphaned keys, placeholder mismatches, length limits, glossary alignment, and JSON syntax. Use it after translation work or in CI before merging.

Protocol Verification checks epistemic protocol definitions for consistency before commit. Use it to run static checks and expert review when validating definitions, soundness, or pre-commit quality.

Review Loop runs a convergence-paced review-resolve cycle over a code change and its governing surfaces. It verifies findings against the codebase and the review base, then re-reviews until each finding is disposed of or left as residual.

Type Realization measures whether a named protocol’s formal contract is realized during an actual run. Use it to test runtime behavior like phase transitions, gates, and stop/proceed outcomes rather than static file quality.

WIO helps choose what to test next, write focused tests, generate realistic workloads, review test value, and diagnose test-suite health. Use it when you need a concrete testing workflow artifact rather than general QA advice.

Enforces citation requirements, source grounding, and 'I don't know' behavior to reduce hallucination. Toggle on for research that needs factual accuracy.

Fact-check and hype-audit any content: video, article, tweet, PDF. Extracts claims, verifies each via web search, and produces a report card with per-claim verdicts and an overall BS score (0-10). Use to evaluate credibility or debunk claims.

Analyze news coverage to distinguish between many independent sources and a single source reprinted many times. Queries GDELT for articles, groups by headline similarity, and estimates the number of distinct origins behind a claim.

Conducts a read-only audit of a Hermes agent profile, comparing declared role to actual tools, skills, access, and behavior, to identify misalignments without automatic rewriting. Produces an evidence-based improvement plan.

A read-only audit tool for installed Hermes skills that identifies overlaps, staleness, broken references, and usage integrity issues without making changes.

Safely consolidate, restructure, deprecate, split, or share references across installed Hermes skills after overlap is established. Generates a read-only plan with rollback snapshots, staged writes, and post-change verification before any mutation.