Local deterministic MCP evidence and release gates with portable Agent Skills
Dailyaiagents/daily-ai-agent-toolkit
Skill Claude CodeCodex
Verify that declared deliverable files exist, are non-empty, remain inside an allowed root, and avoid declared placeholder terms.
Dailyaiagents/daily-ai-agent-toolkit
Skill Claude CodeCodex
Refuse claim records that lack an explicit source or an honest unverified state.
Dailyaiagents/daily-ai-agent-toolkit
Skill Claude CodeCodex
Persist ordered completion steps and refuse skipped, repeated, or out-of-order completion claims.
Dailyaiagents/daily-ai-agent-toolkit
Skill Claude CodeCodex
Compare a JSON output contract with retained evidence files under one workspace root.
Dailyaiagents/daily-ai-agent-toolkit
Skill Claude CodeCodex
Convert a failed dependency or tool step into an explicit blocker record without substituting a narrative success.
Dailyaiagents/daily-ai-agent-toolkit
Skill Claude CodeCodex
Run a labeled fixture set through the bundled deterministic subject and report catch rate and false-refusal rate without converting synthetic performance into production proof.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: