# Send Intent Align to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "intent-align", "name": "Intent Align", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/oscraters/intent-align", "canonicalUrl": "https://clawhub.ai/oscraters/intent-align", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/intent-align", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=intent-align", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "SKILL.md", "assets/mermaid-spec-template.md", "references/core-contract.md", "references/alignment-hub-template.md", "references/adapters/adapter-template.md", "references/adapters/local-repo.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "intent-align", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-05-01T23:39:31.691Z", "expiresAt": "2026-05-08T23:39:31.691Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=intent-align", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=intent-align", "contentDisposition": "attachment; filename=\"intent-align-2.0.0.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "intent-align" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/intent-align" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/intent-align", "downloadUrl": "https://openagent3.xyz/downloads/intent-align", "agentUrl": "https://openagent3.xyz/skills/intent-align/agent", "manifestUrl": "https://openagent3.xyz/skills/intent-align/agent.json", "briefUrl": "https://openagent3.xyz/skills/intent-align/agent.md" } } ``` ## Documentation ### Intent-Align v2 Core Keep execution aligned to user intent while preserving agent autonomy. ### Quick Start Read references/core-contract.md. Create alignment hub from references/alignment-hub-template.md. Run Intent Quality Gate before planning. Select adapters from references/adapters/. Execute phases with realignment and verification gates. ### Workflow Contract Build intent_snapshot and run intent_lint (see core contract). Ask for strictness mode and scope overrides (project/repo/workflow/task). Resolve effective strictness by precedence: task > workflow > repo > project > default. Evaluate ambiguity action using severity + strictness + risk class. Generate phase plan and Mermaid diagram with explicit dependencies and gates. Bind available adapters through capability_matrix. Execute phase-by-phase. Before phase start, run Pre-Execution Clarification Gate. On each phase end, run verification gates (including output conformance) and update drift evidence. If intent or constraints change, apply intent_delta and re-plan only impacted phases. Close with final alignment report and open ambiguity list (if any). ### Autonomy Levels 1 Strict: Require user confirmation before each phase start. 2 Balanced: Require user confirmation at phase end or any critical drift. 3 Aggressive: Auto-continue on low drift; require confirmation on major deltas. 4 Exploratory: Continue with log-only check-ins unless risk or ambiguity threshold is exceeded. Override rule: high-risk ambiguity is never advisory-only; enforce at least soft_gate. Strictness rule: strictness mode controls whether ambiguities block or proceed with guardrails. ### Required References references/core-contract.md references/alignment-hub-template.md references/realignment-protocol.md references/verification-gates.md references/capability-taxonomy.md ### Adapter Selection Use only adapters needed for the task: references/adapters/github.md references/adapters/local-repo.md references/adapters/tracker-generic.md references/adapters/adapter-template.md If no adapter can satisfy a required capability: Generate an ad-hoc adapter spec from adapter-template.md. Add provenance metadata (created_by, created_at, environment_assumptions, tool_access_required). Validate required fields before use. Register the new adapter in capability_matrix.adapters_selected. Continue in degraded mode only if validation fails or auth/capability remains unavailable. ### Anti-Bloat Rules Keep core contract tool-agnostic. Do not add tracker- or host-specific logic to core files. Add a new adapter only for a proven capability gap. Keep schemas single-source; do not duplicate fields across files. Tie each new feature to one concrete failure mode and one test scenario. Generate ad-hoc adapters only for current task scope; do not pre-generate broad catalogs. Use canonical capability IDs first; extend only when needed via namespaced custom IDs. ### Edge Cases Multi-repo: maintain one hub with per-repo adapter bindings and dependency graph. Non-git prototype: use local artifacts and explicit acceptance criteria checkpoints. Team swarm: assign owner per phase and keep decision log in hub. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: oscraters - Version: 2.0.0 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-05-01T23:39:31.691Z - Expires at: 2026-05-08T23:39:31.691Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/intent-align) - [Send to Agent page](https://openagent3.xyz/skills/intent-align/agent) - [JSON manifest](https://openagent3.xyz/skills/intent-align/agent.json) - [Markdown brief](https://openagent3.xyz/skills/intent-align/agent.md) - [Download page](https://openagent3.xyz/downloads/intent-align)