the governed runtime for agent skill workflows, off the leash but on the record
This repository also configures its own agents. See what runx tells them →
Skill Claude CodeCodex
Detect and remove personal data before content crosses a trust boundary, returning usable scrubbed content only when a deterministic residual scan passes. Use for exports, prompts, logs, support material, or outbound handoffs that need a pass, review, or block verdict; it does not move content or handle credentials.
Skill Claude CodeCodex
Internal reflection stage that groups durable run signals by skill and emits validated handoffs to skill-lab improve when evidence clears configured floors. Use diagnose-skill-run, run-history, or skill-lab directly instead.
Skill Claude CodeCodex
Refund one sealed payment through Runx Hosted with bounded authority, approval, idempotency, and readback.
Skill Claude CodeCodex
Prepare, approve, execute, and independently verify a versioned release through a project-owned exact-command profile.
Skill Claude CodeCodex
Produce bounded, source-backed research packets for product, ecosystem, and operator decisions.
Skill Claude CodeCodex
Internal evidence-index analysis and verification stage for the canonical research skill.
Skill Claude CodeCodex
Inspect, safely test, and assess one Runx skill package for capability, trust, and operator readiness. Use when deciding whether to adopt, improve, reject, install, or publish a skill; its evidence-only assess runner is available when native test evidence already exists.
Skill Claude CodeCodex
Read Runx's native receipt history and skill catalog, then return deterministic run outcomes, catalog test coverage, and governance follow-ups without model-authored metrics.
Skill Claude CodeCodex
Derive a CycloneDX software bill of materials from a supplied lockfile deterministically, digest-bound to the exact bytes, with license risks surfaced and unsupported or unpinned lockfiles refused.
Skill Claude CodeCodex
Check a proposed API or data contract against the current contract, real sample payloads, and a compatibility policy, and gate publication behind a refusable verdict with native digest evidence.
Skill Claude CodeCodex
Send one digest-bound provider-neutral message or campaign action through a compatible connector, with an explicit plan runner and approval plus readback at the live-delivery boundary.
Skill Claude CodeCodex
Validate and settle one invoice through Runx Hosted with approval and provider readback.
Skill Claude CodeCodex
Bind an off-runtime action claim to opaque evidence references in a signed Runx receipt without pretending Runx verified the external action.
Skill Claude CodeCodex
Canonical Runx skill-authoring implementation. Use for designing, creating, updating, improving, or adding harness coverage to a Runx skill package; it combines bounded agent judgment with native file writes, inspection, and safe harness validation. When a host skill-creator also triggers, use its general guidance but…
Skill Claude CodeCodex
Plan a digest-bound Slack notification, then deliver the exact approved channel post through any compatible Slack binding with provider readback.
Skill Claude CodeCodex
Read bounded Slack search and thread evidence, plan an exact reply, and deliver an approved reply through any compatible Slack binding with stable-message readback.
Skill Claude CodeCodex
Generate documentation for a project using Sourcey.
Skill Claude CodeCodex
Execute one governed outbound payment through Runx Hosted with approval and provider readback.
Skill Claude CodeCodex
Answer a bounded data question from one declared governed read; use the explicit plan runner when only a schema-validated analysis plan is wanted.
Skill Claude CodeCodex
Execute a Stripe SPT payment through the canonical hosted spend contract.
Skill Claude CodeCodex
Build a scoped taste profile packet from examples, preferences, and explicit dislikes so downstream agents can make style decisions without inventing the user's taste.
Skill Claude CodeCodex
Govern X (Twitter) account work through three lanes: evidence reads, typed act plans for posting and account hygiene, and gated execution with per-act provider evidence. Nothing reaches a live timeline or mutates an account without an explicit approval recorded in the receipt.
Skill Claude CodeCodex
Plan or execute a scoped, time-bounded vault-unseal request through any compatible vault binding, returning only opaque handle metadata and provider readback.
Skill Claude CodeCodex
Prepare an evidence-bound vulnerability publication and publish the exact approved advisory through any compatible provider binding with independent readback.
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: