An AI writing skill that preserves meaning and learns your voice across Claude Code, Codex, Cursor, Gemini, Copilot and OpenCode.
Agent
Fails a draft before it ships. Invoked by humanise Step 6 (external pieces) and Step D5 (mandatory for long-form), and by the self-harness acceptance gate for high-stakes changes. Runs in a separate context with no access to the drafting rationale, because the writer shares the draft's blind spots.
Agent
Draft generator for the benchmark and indistinguishability stages of /humanise improve. Runs in one of two modes set by the spawn prompt, skill (full workflow) or baseline (no skill context), and writes the finished draft to a given output path.
Agent
Checks a draft against its brief or source before it ships: are the specific claims true and supported, and does the piece still say what the brief asked? Invoked by humanise Step 6 and Step D5 for external pieces that carry checkable claims, when a brief or source exists. Unlike the adversarial-reviewer, it receives…
Agent
Drafts bounded engine-change proposals from mined weakness clusters (candidates.json), in the four-part self-harness schema. Separate context; sees only the evidence and the named engine surfaces, with no access to the conversation that produced them.
Agent
Blind pairwise judge for the indistinguishability test (evals/indistinguishability.md). Given two texts, decides which is the real person's and names the deciding signals. No corpus, no conversation, no labels.