Plugin Claude Code
Make No Mistakes — the verification harness for AI coding agents. No unverified work passes.
Plugin Claude Code
Make No Mistakes — the verification harness for AI coding agents. No unverified work passes.
Plugin Claude Code
Make AI coding agents prove their work — frozen specs, budgeted verification loops, hard gates, an independent blind verifier. No unverified work passes.
Agent
Aggregates the refuter panel's independently-committed verdicts into one per-criterion verdict, under the independence guard. Never itself judges the diff. Invoked only by mnm-verify, after every lens for the current round has already committed.
Agent
H/C-tier-only adversarial/misuse lens — injection, authz bypass, unsafe input handling, data-integrity races within the diff's actual touch surface. Read-only. Invoked only by mnm-verify, only at H/C tier.
Agent
Breaks a diff on pure functional correctness — wrong output, crash, violated invariant, uncovered edge case. Read-only, re-runs the decisive check itself. Invoked only by mnm-verify; never by the user, never by the implementer.
Agent
Breaks a diff on regression — anything that worked before this change and doesn't now, including quietly weakened checks. Read-only, re-runs the pass-to-pass suite itself. Invoked only by mnm-verify; never by the user, never by the implementer.
Agent
Reproduction-first lens for bug-fix DoDs only — the original bug must be reproduced against pre-fix code before any fix claim is trusted. Read-only. Invoked only by mnm-verify, only when the DoD's Goal is a bug fix.
Agent
Breaks a diff on spec-conformance — DoD criteria met in letter but not substance, Non-Goals silently crossed, write-scope exceeded. Read-only. Invoked only by mnm-verify; never by the user, never by the implementer.
Hook
Runs before the agent uses a tool for Write, Edit, MultiEdit, Read and Bash tool calls, executing pretooluse-protect.sh. From momomuchu/make-no-mistakes.
Hook
Runs when the agent finishes a response, executing stop-loop.sh. From momomuchu/make-no-mistakes.
Skill Claude CodeCodex
Zero-unverified-work execution harness for coding agents. Use when the user invokes /make-no-mistakes, says "make no mistakes", "don't get this wrong", "this must be correct", "prove it works", or hands over work where a wrong result is costly. Turns any task into a verified pipeline: freeze what "correct" means…
Skill Claude CodeCodex
Internal creation-side specialist for /make-no-mistakes. Invoked ONLY by the orchestrator's own Skill-tool call (disable-model-invocation: true), and only for H/C-tier tournaments, and only after the arbiter has declared a winner AND the verification panel's own citations name a specific way a losing candidate handled…
Skill Claude CodeCodex
Internal creation-side specialist for /make-no-mistakes. Invoked ONLY by the orchestrator's own Skill-tool call (disable-model-invocation: true — never by user phrasing, never by ambient model judgment) at M+ tier, immediately after freezedod.sh succeeds and before any Implementer runs. Reads the frozen DoD and the…
Skill Claude CodeCodex
Internal creation-side specialist for /make-no-mistakes. Invoked ONLY by the orchestrator's own Skill-tool call (disable-model-invocation: true) at M+ tier, in parallel with mnm-planner, immediately after freezedod.sh succeeds. Authors falsifying tests for every acceptance criterion in the frozen DoD, structurally…
Skill Claude CodeCodex
Adversarial verification panel for make-no-mistakes. Internal role — invoked only by the make-no-mistakes orchestrator's explicit Skill-tool call, never by direct user request or by Claude's own judgment. Certifies one diff against one frozen dod.lock.md through independent, break-it-first refuter lenses, then…