[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78MAC2FA5672VP93MKFA2Y3":3,"skill-same:01M78MAC2FA5672VP93MKFA2Y3":45},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":28,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M78MAC2FA5672VP93MKFA2Y3","secure-execution-review",null,"Check MVM runtime, backend, policy, networking, package, or lifecycle changes against local authority and evidenced security guarantees. Use before designing or reviewing security-sensitive MVM work.","tinylabscom\u002Fmvm","tinylabscom","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F122397106?v=4","01M6CJPHHRE2TKXMAH0Z27PC0B","01M6CJPHP46MATSCYX3D0TDGP8",47,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Ftinylabscom\u002Fmvm\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsecure-execution-review\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: secure-execution-review\ndescription: Check MVM runtime, backend, policy, networking, package, or lifecycle changes against local authority and evidenced security guarantees. Use before designing or reviewing security-sensitive MVM work.\n---\n\n# Review a secure execution change\n\nRead `AGENTS.md` first. Start from the governing issue and the ADRs and\ncontracts it links. Use Trail to locate the owning production path and its\ncallers. ADR-001's claims table and backend matrix govern shipped security\nclaims; proposals, fixtures, and tests for other backends do not extend them.\nDo not use this skill to turn a small unrelated change into a repository-wide\naudit. For a focused security review, use the security-audit skill in guidance\nmode when available; a full codebase audit requires its own request. Rust\nchanges still follow the repository's Rust-skill rule.\n\n1. **State the contract.** Identify the user-visible operation, selected\n   backend, caller, signed plan or pack, trust boundary, and exact outcome.\n   Classify each involved capability as shipped, preview, or proposed.\n   If ownership, public API, authorization, or the security model changes,\n   obtain the required design\u002FADR review before implementation.\n2. **Trace the one authority path.** Follow the operation from the client\n   through policy resolution, signed admission, the selected backend, and\n   host-side enforcement. For network or mailbox work, distinguish canonical\n   MVM local contracts from optional MVMD cross-host delivery. Check that\n   each backend either enforces the advertised capability or refuses it\n   before execution; do not introduce a second gate, guest NIC, SSH path,\n   silent fallback, or cache-based authorization.\n   When platforms genuinely implement the same capability differently,\n   keep the caller on one existing trait or contract and put the implementations\n   under `platforms\u002F`, selecting modules at their boundary rather than scattering\n   `cfg(unix)` through callers. Preserve unsupported-platform refusals; do not\n   invent a trait for a single-platform module, a test assertion, or one syscall,\n   or hide module declarations behind `include!`. Keep Windows\u002FWHP as a future\n   target of the same contract, not a current enforcement claim: a Windows\n   adapter needs its own owner-only storage, isolation and refusal witnesses\n   before it can replace a fail-closed platform module.\n3. **Name the threat and evidence.** Say which actor may be compromised and\n   whether protected data is at rest, in transit, in use, or retained after\n   termination. Check issuer, digest, revocation freshness, epoch, replay,\n   bounds, key custody, and audit where relevant. Do not call a host-held\n   key, best-effort RAM scrub, or simulated exchange a malicious-host or\n   physical-erasure guarantee. State any hardware requirement explicitly.\n4. **Review changed transitions.** For launch, build\u002Fcutover, mailbox,\n   offline\u002Freconnect, quarantine, replacement, or pack admission, enumerate\n   states, transition owner, durable evidence, deadlines, cancellation,\n   crash recovery, and typed refusal or unknown outcomes. A verified build\n   is not an admitted or ready replacement; mailbox acceptance, delivery,\n   and application acknowledgement are separate.\n5. **Prove behavior at the right boundary.** Add a happy-path witness and\n   negative tests for the relevant tampering, wrong principal\u002Fkey, stale\n   epoch, replay, unavailable enforcement, quota exhaustion, and cleanup\n   failure. Use the actual backend or label mock\u002Ffixture evidence as such.\n   Measure cold and warm readiness with admission included, and record\n   image preparation, launch, and safe replacement separately when touched.\n   Do not trade a required check for a faster reported clock.\n6. **Report honestly.** Update the owning issue, applicable contract and\n   user documentation with observed status and links to tests. Run the\n   repository's scoped and required gates, name exact source revisions and\n   commands, and call out unrun live\u002Fhardware checks. Keep unfinished\n   checkboxes open. Do not claim another repository's design direction is\n   implemented here.\n","# Review a secure execution change\n\nRead `AGENTS.md` first. Start from the governing issue and the ADRs and\ncontracts it links. Use Trail to locate the owning production path and its\ncallers. ADR-001's claims table and backend matrix govern shipped security\nclaims; proposals, fixtures, and tests for other backends do not extend them.\nDo not use this skill to turn a small unrelated change into a repository-wide\naudit. For a focused security review, use the security-audit skill in guidance\nmode when available; a full codebase audit requires its own request. Rust\nchanges still follow the repository's Rust-skill rule.\n\n1. **State the contract.** Identify the user-visible operation, selected\n   backend, caller, signed plan or pack, trust boundary, and exact outcome.\n   Classify each involved capability as shipped, preview, or proposed.\n   If ownership, public API, authorization, or the security model changes,\n   obtain the required design\u002FADR review before implementation.\n2. **Trace the one authority path.** Follow the operation from the client\n   through policy resolution, signed admission, the selected backend, and\n   host-side enforcement. For network or mailbox work, distinguish canonical\n   MVM local contracts from optional MVMD cross-host delivery. Check that\n   each backend either enforces the advertised capability or refuses it\n   before execution; do not introduce a second gate, guest NIC, SSH path,\n   silent fallback, or cache-based authorization.\n   When platforms genuinely implement the same capability differently,\n   keep the caller on one existing trait or contract and put the implementations\n   under `platforms\u002F`, selecting modules at their boundary rather than scattering\n   `cfg(unix)` through callers. Preserve unsupported-platform refusals; do not\n   invent a trait for a single-platform module, a test assertion, or one syscall,\n   or hide module declarations behind `include!`. Keep Windows\u002FWHP as a future\n   target of the same contract, not a current enforcement claim: a Windows\n   adapter needs its own owner-only storage, isolation and refusal witnesses\n   before it can replace a fail-closed platform module.\n3. **Name the threat and evidence.** Say which actor may be compromised and\n   whether protected data is at rest, in transit, in use, or retained after\n   termination. Check issuer, digest, revocation freshness, epoch, replay,\n   bounds, key custody, and audit where relevant. Do not call a host-held\n   key, best-effort RAM scrub, or simulated exchange a malicious-host or\n   physical-erasure guarantee. State any hardware requirement explicitly.\n4. **Review changed transitions.** For launch, build\u002Fcutover, mailbox,\n   offline\u002Freconnect, quarantine, replacement, or pack admission, enumerate\n   states, transition owner, durable evidence, deadlines, cancellation,\n   crash recovery, and typed refusal or unknown outcomes. A verified build\n   is not an admitted or ready replacement; mailbox acceptance, delivery,\n   and application acknowledgement are separate.\n5. **Prove behavior at the right boundary.** Add a happy-path witness and\n   negative tests for the relevant tampering, wrong principal\u002Fkey, stale\n   epoch, replay, unavailable enforcement, quota exhaustion, and cleanup\n   failure. Use the actual backend or label mock\u002Ffixture evidence as such.\n   Measure cold and warm readiness with admission included, and record\n   image preparation, launch, and safe replacement separately when touched.\n   Do not trade a required check for a faster reported clock.\n6. **Report honestly.** Update the owning issue, applicable contract and\n   user documentation with observed status and links to tests. Run the\n   repository's scoped and required gates, name exact source revisions and\n   commands, and call out unrun live\u002Fhardware checks. Keep unfinished\n   checkboxes open. Do not claim another repository's design direction is\n   implemented here.\n",{"name":5,"description":7},"main","5b25b3d870f83a427a8aa24fc819e00fe42a072d",".agents\u002Fskills\u002Fsecure-execution-review",6,"https:\u002F\u002Fgithub.com\u002Ftinylabscom\u002Fmvm","https:\u002F\u002Frunmvm.com\u002F","https:\u002F\u002Fgithub.com\u002Ftinylabscom","2026-11-15 03:39:40",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add tinylabscom\u002Fmvm --skill secure-execution-review",{"tool":37,"command":38},"bunx","bunx skills add tinylabscom\u002Fmvm --skill secure-execution-review",{"tool":40,"command":41},"pnpm","pnpm dlx skills add tinylabscom\u002Fmvm --skill secure-execution-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftinylabscom\u002Fmvm\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsecure-execution-review\u002FSKILL.md",true,[],[]]