[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6X1KETSPTV22RWZYAQ711TC":3,"skill-same:01M6X1KETSPTV22RWZYAQ711TC":94},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":45,"linkOnly":46,"markdownContent":47,"markdownBody":48,"frontMatter":49,"refName":70,"commitSha":71,"skillDirectory":72,"forksCount":73,"repositoryHtmlUrl":74,"repositoryHomepageUrl":75,"ownerHtmlUrl":76,"ownerBlogUrl":77,"ownerDescription":78,"updatedAt":79,"install":80,"resources":93},"01M6X1KETSPTV22RWZYAQ711TC","external-comms",null,"Workflow for scanning, drafting, and presenting community responses with a human review gate","github\u002Fsquad","github","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F9919?v=4","01M6AC83DS133TPXFSP1DHNCC6","GitHub","01M6X0MZ6B0WYMMDPK3Q61KER6",3391,"MIT","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fpackages\u002Fsquad-sdk\u002Ftemplates\u002Fskills\u002Fexternal-comms\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"external comms","comms","drafting","cli","mcp-server","server","communications","review","api","candidate","issue list","workflow","when","workflows","MCP","open","labels","issues","search","community","coverage","2026-11-10 10:12:30",false,"---\r\nname: \"external-comms\"\r\ndescription: \"Workflow for scanning, drafting, and presenting community responses with a human review gate\"\r\ndomain: \"community, communication, workflow\"\r\nconfidence: \"low\"\r\nsource: \"manual (external communications workflow)\"\r\ntools:\r\n  - name: \"github-mcp-server-list_issues\"\r\n    description: \"List open issues for scan candidates and lightweight triage\"\r\n    when: \"Use for recent open issue scans before thread-level review\"\r\n  - name: \"github-mcp-server-issue_read\"\r\n    description: \"Read the full issue, comments, and labels before drafting\"\r\n    when: \"Use after selecting a candidate so the communications owner has complete thread context\"\r\n  - name: \"github-mcp-server-search_issues\"\r\n    description: \"Search for candidate issues or prior squad responses\"\r\n    when: \"Use when filtering by keywords, labels, or duplicate response checks\"\r\n  - name: \"gh CLI\"\r\n    description: \"Fallback for GitHub issue comments and discussions workflows\"\r\n    when: \"Use gh issue list\u002Fcomment and gh api or gh api graphql when MCP coverage is incomplete\"\r\n---\r\n\r\n## Context\r\n\r\nPhase 1 is **draft-only mode**.\r\n\r\n- The communications owner scans issues and discussions, drafts responses with the humanizer skill, and presents a review table for human approval.\r\n- **Human review gate is mandatory** — the communications owner never posts autonomously.\r\n- Every action is logged to `.squad\u002Fcomms\u002Faudit\u002F`.\r\n- This workflow is triggered manually only (\"check community\") — no automated activation in Phase 1.\r\n\r\n## Patterns\r\n\r\n### 1. Scan\r\n\r\nFind unanswered community items with GitHub MCP tools first, or `gh issue list` \u002F `gh api` as fallback for issues and discussions.\r\n\r\n- Include **open** issues and discussions only.\r\n- Filter for items with **no squad team response**.\r\n- Limit to items created in the last 7 days.\r\n- Exclude items labeled `squad:internal` or `wontfix`.\r\n- Include discussions **and** issues in the same sweep.\r\n- Phase 1 scope is **issues and discussions only** — do not draft PR replies.\r\n\r\n### Discussion Handling (Phase 1)\r\n\r\nDiscussions use the GitHub Discussions API, which differs from issues:\r\n\r\n- **Scan:** `gh api \u002Frepos\u002F{owner}\u002F{repo}\u002Fdiscussions --jq '.[] | select(.answer_chosen_at == null)'` to find unanswered discussions\r\n- **Categories:** Filter by Q&A and General categories only (skip Announcements, Show and Tell)\r\n- **Answers vs comments:** In Q&A discussions, the communications owner drafts an \"answer\" (not a comment). The human marks it as accepted answer after posting.\r\n- **Phase 1 scope:** Issues and Discussions ONLY. No PR comments.\r\n\r\n### 2. Classify\r\n\r\nDetermine the response type before drafting.\r\n\r\n- Welcome (new contributor)\r\n- Troubleshooting (bug\u002Fhelp)\r\n- Feature guidance (feature request\u002Fhow-to)\r\n- Redirect (wrong repo\u002Fscope)\r\n- Acknowledgment (confirmed, no fix)\r\n- Closing (resolved)\r\n- Technical uncertainty (unknown cause)\r\n- Empathetic disagreement (pushback on a decision or design)\r\n- Information request (need more reproduction details or context)\r\n\r\n### Template Selection Guide\r\n\r\n| Signal in Issue\u002FDiscussion | → Response Type | Template |\r\n|---------------------------|-----------------|----------|\r\n| New contributor (0 prior issues) | Welcome | T1 |\r\n| Error message, stack trace, \"doesn't work\" | Troubleshooting | T2 |\r\n| \"How do I...?\", \"Can Squad...?\", \"Is there a way to...?\" | Feature Guidance | T3 |\r\n| Wrong repo, out of scope for Squad | Redirect | T4 |\r\n| Confirmed bug, no fix available yet | Acknowledgment | T5 |\r\n| Fix shipped, PR merged that resolves issue | Closing | T6 |\r\n| Unclear cause, needs investigation | Technical Uncertainty | T7 |\r\n| Author disagrees with a decision or design | Empathetic Disagreement | T8 |\r\n| Need more reproduction info or context | Information Request | T9 |\r\n\r\nUse exactly one template as the base draft. Replace placeholders with issue-specific details, then apply the humanizer patterns. If the thread spans multiple signals, choose the highest-risk template and capture the nuance in the thread summary.\r\n\r\n### Confidence Classification\r\n\r\n| Confidence | Criteria | Example |\r\n|-----------|----------|---------|\r\n| 🟢 High | Answer exists in Squad docs or FAQ, similar question answered before, no technical ambiguity | \"How do I install Squad?\" |\r\n| 🟡 Medium | Technical answer is sound but involves judgment calls, OR docs exist but don't perfectly match the question, OR tone is tricky | \"Can Squad work with Azure DevOps?\" (yes, but setup is nuanced) |\r\n| 🔴 Needs Review | Technical uncertainty, policy\u002Froadmap question, potential reputational risk, author is frustrated\u002Fangry, question about unreleased features | \"When will Squad support Claude?\" |\r\n\r\n**Auto-escalation rules:**\r\n- Any mention of competitors → 🔴\r\n- Any mention of pricing\u002Flicensing → 🔴\r\n- Author has >3 follow-up comments without resolution → 🔴\r\n- Question references a closed-wontfix issue → 🔴\r\n\r\n### 3. Draft\r\n\r\nUse the humanizer skill for every draft.\r\n\r\n- Complete **Thread-Read Verification** before writing.\r\n- Read the **full thread**, including all comments, before writing.\r\n- Select the matching template from the **Template Selection Guide** and record the template ID in the review notes.\r\n- Treat templates as reusable drafting assets: keep the structure, replace placeholders, and only improvise when the thread truly requires it.\r\n- Validate the draft against the humanizer anti-patterns.\r\n- Flag long threads (`>10` comments) with `⚠️`.\r\n\r\n### Thread-Read Verification\r\n\r\nBefore drafting, the communications owner MUST verify complete thread coverage:\r\n\r\n1. **Count verification:** Compare API comment count with actually-read comments. If mismatch, abort draft.\r\n2. **Deleted comment check:** Use `gh api` timeline to detect deleted comments. If found, flag as ⚠️ in review table.\r\n3. **Thread summary:** Include in every draft: \"Thread: {N} comments, last activity {date}, {summary of key points}\"\r\n4. **Long thread flag:** If >10 comments, add ⚠️ to review table and include condensed thread summary\r\n5. **Evidence line in review table:** Each draft row includes \"Read: {N}\u002F{total} comments\" column\r\n\r\n### 4. Present\r\n\r\nShow drafts for review in this exact format:\r\n\r\n```text\r\n📝 Community Response Drafts\r\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\r\n\r\n| # | Item | Author | Type | Confidence | Read | Preview |\r\n|---|------|--------|------|------------|------|---------|\r\n| 1 | Issue #N | @user | Type | 🟢\u002F🟡\u002F🔴 | N\u002FN | \"First words...\" |\r\n\r\nConfidence: 🟢 High | 🟡 Medium | 🔴 Needs review\r\n\r\nFull drafts below ▼\r\n```\r\n\r\nEach full draft must begin with the thread summary line:\r\n`Thread: {N} comments, last activity {date}, {summary of key points}`\r\n\r\n### 5. Human Action\r\n\r\nWait for explicit human direction before anything is posted.\r\n\r\n- `comms approve 1 3` — approve drafts 1 and 3\r\n- `comms edit 2` — edit draft 2\r\n- `comms skip` — skip all\r\n- `banana` — freeze all pending (safe word)\r\n\r\n### Rollback — Bad Post Recovery\r\n\r\nIf a posted response turns out to be wrong, inappropriate, or needs correction:\r\n\r\n1. **Delete the comment:**\r\n   - Issues: `gh api -X DELETE \u002Frepos\u002F{owner}\u002F{repo}\u002Fissues\u002Fcomments\u002F{comment_id}`\r\n   - Discussions: `gh api graphql -f query='mutation { deleteDiscussionComment(input: {id: \"{node_id}\"}) { comment { id } } }'`\r\n2. **Log the deletion:** Write audit entry with action `delete`, include reason and original content\r\n3. **Draft replacement** (if needed): The communications owner drafts a corrected response and repeats the normal review cycle\r\n4. **Postmortem:** If the error reveals a pattern gap, update humanizer anti-patterns or add a new test case\r\n\r\n**Safe word — `banana`:**\r\n- Immediately freezes all pending drafts in the review queue\r\n- No new scans or drafts until `comms resume` is issued\r\n- Audit entry logged with halter identity and reason\r\n\r\n### 6. Post\r\n\r\nAfter approval:\r\n\r\n- Human posts via `gh issue comment` for issues or `gh api` for discussion answers\u002Fcomments.\r\n- The communications owner helps by preparing the CLI command.\r\n- Write the audit entry after the posting action.\r\n\r\n### 7. Audit\r\n\r\nLog every action.\r\n\r\n- Location: `.squad\u002Fcomms\u002Faudit\u002F{timestamp}.md`\r\n- Required fields vary by action — see `.squad\u002Fcomms\u002Ftemplates\u002Faudit-entry.md` Conditional Fields table\r\n- Universal required fields: `timestamp`, `action`\r\n- All other fields are conditional on the action type\r\n\r\n## Examples\r\n\r\nThese are reusable templates. Keep the structure, replace placeholders, and adjust only where the thread requires it.\r\n\r\n### Example scan command\r\n\r\n```bash\r\ngh issue list --state open --json number,title,author,labels,comments --limit 20\r\n```\r\n\r\n### Example review table\r\n\r\n```text\r\n📝 Community Response Drafts\r\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\r\n\r\n| # | Item | Author | Type | Confidence | Read | Preview |\r\n|---|------|--------|------|------------|------|---------|\r\n| 1 | Issue #426 | @newdev | Welcome | 🟢 | 1\u002F1 | \"Hey @newdev! Welcome to Squad...\" |\r\n| 2 | Discussion #18 | @builder | Feature guidance | 🟡 | 4\u002F4 | \"Great question! Today the CLI...\" |\r\n| 3 | Issue #431 ⚠️ | @debugger | Technical uncertainty | 🔴 | 12\u002F12 | \"Interesting find, @debugger...\" |\r\n\r\nConfidence: 🟢 High | 🟡 Medium | 🔴 Needs review\r\n\r\nFull drafts below ▼\r\n```\r\n\r\n### Example audit entry (post action)\r\n\r\n```markdown\r\n---\r\ntimestamp: \"2026-03-16T21:30:00Z\"\r\naction: \"post\"\r\nitem_number: 426\r\ndraft_id: 1\r\nreviewer: \"@bradygaster\"\r\n---\r\n\r\n## Context (draft, approve, edit, skip, post, delete actions)\r\n- Thread depth: 3\r\n- Response type: welcome\r\n- Confidence: 🟢\r\n- Long thread flag: false\r\n\r\n## Draft Content (draft, edit, post actions)\r\nThread: 3 comments, last activity 2026-03-16, reporter hit a preview-build regression after install.\r\n\r\nHey @newdev! Welcome to Squad 👋 Thanks for opening this.\r\nWe reproduced the issue in preview builds and we're checking the regression point now.\r\nLet us know if you can share the command you ran right before the failure.\r\n\r\n## Post Result (post, delete actions)\r\nhttps:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fissues\u002F426#issuecomment-123456\r\n```\r\n\r\n### T1 — Welcome\r\n\r\n```text\r\nHey {author}! Welcome to Squad 👋 Thanks for opening this.\r\n{specific acknowledgment or first answer}\r\nLet us know if you have questions — happy to help!\r\n```\r\n\r\n### T2 — Troubleshooting\r\n\r\n```text\r\nThanks for the detailed report, {author}!\r\nHere's what we think is happening: {explanation}\r\n{steps or workaround}\r\nLet us know if that helps, or if you're seeing something different.\r\n```\r\n\r\n### T3 — Feature Guidance\r\n\r\n```text\r\nGreat question! {context on current state}\r\n{guidance or workaround}\r\nWe've noted this as a potential improvement — {tracking info if applicable}.\r\n```\r\n\r\n### T4 — Redirect\r\n\r\n```text\r\nThanks for reaching out! This one is actually better suited for {correct location}.\r\n{brief explanation of why}\r\nFeel free to open it there — they'll be able to help!\r\n```\r\n\r\n### T5 — Acknowledgment\r\n\r\n```text\r\nGood catch, {author}. We've confirmed this is a real issue.\r\n{what we know so far}\r\nWe'll update this thread when we have a fix. Thanks for flagging it!\r\n```\r\n\r\n### T6 — Closing\r\n\r\n```text\r\nThis should be resolved in {version\u002FPR}! 🎉\r\n{brief summary of what changed}\r\nThanks for reporting this, {author} — it made Squad better.\r\n```\r\n\r\n### T7 — Technical Uncertainty\r\n\r\n```text\r\nInteresting find, {author}. We're not 100% sure what's causing this yet.\r\nHere's what we've ruled out: {list}\r\nWe'd love more context if you have it — {specific ask}.\r\nWe'll dig deeper and update this thread.\r\n```\r\n\r\n### T8 — Empathetic Disagreement\r\n\r\n```text\r\nWe hear you, {author}. That's a fair concern.\r\n\r\nThe current design choice was driven by {reason}. We know it's not ideal for every use case.\r\n\r\n{what alternatives exist or what trade-off was made}\r\n\r\nIf you have ideas for how to make this work better for your scenario, we'd love to hear them — open a discussion or drop your thoughts here!\r\n```\r\n\r\n### T9 — Information Request\r\n\r\n```text\r\nThanks for reporting this, {author}!\r\n\r\nTo help us dig into this, could you share:\r\n- {specific ask 1}\r\n- {specific ask 2}\r\n- {specific ask 3, if applicable}\r\n\r\nThat context will help us narrow down what's happening. Appreciate it!\r\n```\r\n\r\n## Anti-Patterns\r\n\r\n- ❌ Posting without human review (NEVER — this is the cardinal rule)\r\n- ❌ Drafting without reading full thread (context is everything)\r\n- ❌ Ignoring confidence flags (🔴 items need Lead or human review)\r\n- ❌ Scanning closed issues (only open items)\r\n- ❌ Responding to issues labeled `squad:internal` or `wontfix`\r\n- ❌ Skipping audit logging (every action must be recorded)\r\n- ❌ Drafting for issues where a squad member already responded (avoid duplicates)\r\n- ❌ Drafting pull request responses in Phase 1 (issues\u002Fdiscussions only)\r\n- ❌ Treating templates like loose examples instead of reusable drafting assets\r\n- ❌ Asking for more info without specific requests\r\n","## Context\r\n\r\nPhase 1 is **draft-only mode**.\r\n\r\n- The communications owner scans issues and discussions, drafts responses with the humanizer skill, and presents a review table for human approval.\r\n- **Human review gate is mandatory** — the communications owner never posts autonomously.\r\n- Every action is logged to `.squad\u002Fcomms\u002Faudit\u002F`.\r\n- This workflow is triggered manually only (\"check community\") — no automated activation in Phase 1.\r\n\r\n## Patterns\r\n\r\n### 1. Scan\r\n\r\nFind unanswered community items with GitHub MCP tools first, or `gh issue list` \u002F `gh api` as fallback for issues and discussions.\r\n\r\n- Include **open** issues and discussions only.\r\n- Filter for items with **no squad team response**.\r\n- Limit to items created in the last 7 days.\r\n- Exclude items labeled `squad:internal` or `wontfix`.\r\n- Include discussions **and** issues in the same sweep.\r\n- Phase 1 scope is **issues and discussions only** — do not draft PR replies.\r\n\r\n### Discussion Handling (Phase 1)\r\n\r\nDiscussions use the GitHub Discussions API, which differs from issues:\r\n\r\n- **Scan:** `gh api \u002Frepos\u002F{owner}\u002F{repo}\u002Fdiscussions --jq '.[] | select(.answer_chosen_at == null)'` to find unanswered discussions\r\n- **Categories:** Filter by Q&A and General categories only (skip Announcements, Show and Tell)\r\n- **Answers vs comments:** In Q&A discussions, the communications owner drafts an \"answer\" (not a comment). The human marks it as accepted answer after posting.\r\n- **Phase 1 scope:** Issues and Discussions ONLY. No PR comments.\r\n\r\n### 2. Classify\r\n\r\nDetermine the response type before drafting.\r\n\r\n- Welcome (new contributor)\r\n- Troubleshooting (bug\u002Fhelp)\r\n- Feature guidance (feature request\u002Fhow-to)\r\n- Redirect (wrong repo\u002Fscope)\r\n- Acknowledgment (confirmed, no fix)\r\n- Closing (resolved)\r\n- Technical uncertainty (unknown cause)\r\n- Empathetic disagreement (pushback on a decision or design)\r\n- Information request (need more reproduction details or context)\r\n\r\n### Template Selection Guide\r\n\r\n| Signal in Issue\u002FDiscussion | → Response Type | Template |\r\n|---------------------------|-----------------|----------|\r\n| New contributor (0 prior issues) | Welcome | T1 |\r\n| Error message, stack trace, \"doesn't work\" | Troubleshooting | T2 |\r\n| \"How do I...?\", \"Can Squad...?\", \"Is there a way to...?\" | Feature Guidance | T3 |\r\n| Wrong repo, out of scope for Squad | Redirect | T4 |\r\n| Confirmed bug, no fix available yet | Acknowledgment | T5 |\r\n| Fix shipped, PR merged that resolves issue | Closing | T6 |\r\n| Unclear cause, needs investigation | Technical Uncertainty | T7 |\r\n| Author disagrees with a decision or design | Empathetic Disagreement | T8 |\r\n| Need more reproduction info or context | Information Request | T9 |\r\n\r\nUse exactly one template as the base draft. Replace placeholders with issue-specific details, then apply the humanizer patterns. If the thread spans multiple signals, choose the highest-risk template and capture the nuance in the thread summary.\r\n\r\n### Confidence Classification\r\n\r\n| Confidence | Criteria | Example |\r\n|-----------|----------|---------|\r\n| 🟢 High | Answer exists in Squad docs or FAQ, similar question answered before, no technical ambiguity | \"How do I install Squad?\" |\r\n| 🟡 Medium | Technical answer is sound but involves judgment calls, OR docs exist but don't perfectly match the question, OR tone is tricky | \"Can Squad work with Azure DevOps?\" (yes, but setup is nuanced) |\r\n| 🔴 Needs Review | Technical uncertainty, policy\u002Froadmap question, potential reputational risk, author is frustrated\u002Fangry, question about unreleased features | \"When will Squad support Claude?\" |\r\n\r\n**Auto-escalation rules:**\r\n- Any mention of competitors → 🔴\r\n- Any mention of pricing\u002Flicensing → 🔴\r\n- Author has >3 follow-up comments without resolution → 🔴\r\n- Question references a closed-wontfix issue → 🔴\r\n\r\n### 3. Draft\r\n\r\nUse the humanizer skill for every draft.\r\n\r\n- Complete **Thread-Read Verification** before writing.\r\n- Read the **full thread**, including all comments, before writing.\r\n- Select the matching template from the **Template Selection Guide** and record the template ID in the review notes.\r\n- Treat templates as reusable drafting assets: keep the structure, replace placeholders, and only improvise when the thread truly requires it.\r\n- Validate the draft against the humanizer anti-patterns.\r\n- Flag long threads (`>10` comments) with `⚠️`.\r\n\r\n### Thread-Read Verification\r\n\r\nBefore drafting, the communications owner MUST verify complete thread coverage:\r\n\r\n1. **Count verification:** Compare API comment count with actually-read comments. If mismatch, abort draft.\r\n2. **Deleted comment check:** Use `gh api` timeline to detect deleted comments. If found, flag as ⚠️ in review table.\r\n3. **Thread summary:** Include in every draft: \"Thread: {N} comments, last activity {date}, {summary of key points}\"\r\n4. **Long thread flag:** If >10 comments, add ⚠️ to review table and include condensed thread summary\r\n5. **Evidence line in review table:** Each draft row includes \"Read: {N}\u002F{total} comments\" column\r\n\r\n### 4. Present\r\n\r\nShow drafts for review in this exact format:\r\n\r\n```text\r\n📝 Community Response Drafts\r\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\r\n\r\n| # | Item | Author | Type | Confidence | Read | Preview |\r\n|---|------|--------|------|------------|------|---------|\r\n| 1 | Issue #N | @user | Type | 🟢\u002F🟡\u002F🔴 | N\u002FN | \"First words...\" |\r\n\r\nConfidence: 🟢 High | 🟡 Medium | 🔴 Needs review\r\n\r\nFull drafts below ▼\r\n```\r\n\r\nEach full draft must begin with the thread summary line:\r\n`Thread: {N} comments, last activity {date}, {summary of key points}`\r\n\r\n### 5. Human Action\r\n\r\nWait for explicit human direction before anything is posted.\r\n\r\n- `comms approve 1 3` — approve drafts 1 and 3\r\n- `comms edit 2` — edit draft 2\r\n- `comms skip` — skip all\r\n- `banana` — freeze all pending (safe word)\r\n\r\n### Rollback — Bad Post Recovery\r\n\r\nIf a posted response turns out to be wrong, inappropriate, or needs correction:\r\n\r\n1. **Delete the comment:**\r\n   - Issues: `gh api -X DELETE \u002Frepos\u002F{owner}\u002F{repo}\u002Fissues\u002Fcomments\u002F{comment_id}`\r\n   - Discussions: `gh api graphql -f query='mutation { deleteDiscussionComment(input: {id: \"{node_id}\"}) { comment { id } } }'`\r\n2. **Log the deletion:** Write audit entry with action `delete`, include reason and original content\r\n3. **Draft replacement** (if needed): The communications owner drafts a corrected response and repeats the normal review cycle\r\n4. **Postmortem:** If the error reveals a pattern gap, update humanizer anti-patterns or add a new test case\r\n\r\n**Safe word — `banana`:**\r\n- Immediately freezes all pending drafts in the review queue\r\n- No new scans or drafts until `comms resume` is issued\r\n- Audit entry logged with halter identity and reason\r\n\r\n### 6. Post\r\n\r\nAfter approval:\r\n\r\n- Human posts via `gh issue comment` for issues or `gh api` for discussion answers\u002Fcomments.\r\n- The communications owner helps by preparing the CLI command.\r\n- Write the audit entry after the posting action.\r\n\r\n### 7. Audit\r\n\r\nLog every action.\r\n\r\n- Location: `.squad\u002Fcomms\u002Faudit\u002F{timestamp}.md`\r\n- Required fields vary by action — see `.squad\u002Fcomms\u002Ftemplates\u002Faudit-entry.md` Conditional Fields table\r\n- Universal required fields: `timestamp`, `action`\r\n- All other fields are conditional on the action type\r\n\r\n## Examples\r\n\r\nThese are reusable templates. Keep the structure, replace placeholders, and adjust only where the thread requires it.\r\n\r\n### Example scan command\r\n\r\n```bash\r\ngh issue list --state open --json number,title,author,labels,comments --limit 20\r\n```\r\n\r\n### Example review table\r\n\r\n```text\r\n📝 Community Response Drafts\r\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\r\n\r\n| # | Item | Author | Type | Confidence | Read | Preview |\r\n|---|------|--------|------|------------|------|---------|\r\n| 1 | Issue #426 | @newdev | Welcome | 🟢 | 1\u002F1 | \"Hey @newdev! Welcome to Squad...\" |\r\n| 2 | Discussion #18 | @builder | Feature guidance | 🟡 | 4\u002F4 | \"Great question! Today the CLI...\" |\r\n| 3 | Issue #431 ⚠️ | @debugger | Technical uncertainty | 🔴 | 12\u002F12 | \"Interesting find, @debugger...\" |\r\n\r\nConfidence: 🟢 High | 🟡 Medium | 🔴 Needs review\r\n\r\nFull drafts below ▼\r\n```\r\n\r\n### Example audit entry (post action)\r\n\r\n```markdown\r\n---\r\ntimestamp: \"2026-03-16T21:30:00Z\"\r\naction: \"post\"\r\nitem_number: 426\r\ndraft_id: 1\r\nreviewer: \"@bradygaster\"\r\n---\r\n\r\n## Context (draft, approve, edit, skip, post, delete actions)\r\n- Thread depth: 3\r\n- Response type: welcome\r\n- Confidence: 🟢\r\n- Long thread flag: false\r\n\r\n## Draft Content (draft, edit, post actions)\r\nThread: 3 comments, last activity 2026-03-16, reporter hit a preview-build regression after install.\r\n\r\nHey @newdev! Welcome to Squad 👋 Thanks for opening this.\r\nWe reproduced the issue in preview builds and we're checking the regression point now.\r\nLet us know if you can share the command you ran right before the failure.\r\n\r\n## Post Result (post, delete actions)\r\nhttps:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fissues\u002F426#issuecomment-123456\r\n```\r\n\r\n### T1 — Welcome\r\n\r\n```text\r\nHey {author}! Welcome to Squad 👋 Thanks for opening this.\r\n{specific acknowledgment or first answer}\r\nLet us know if you have questions — happy to help!\r\n```\r\n\r\n### T2 — Troubleshooting\r\n\r\n```text\r\nThanks for the detailed report, {author}!\r\nHere's what we think is happening: {explanation}\r\n{steps or workaround}\r\nLet us know if that helps, or if you're seeing something different.\r\n```\r\n\r\n### T3 — Feature Guidance\r\n\r\n```text\r\nGreat question! {context on current state}\r\n{guidance or workaround}\r\nWe've noted this as a potential improvement — {tracking info if applicable}.\r\n```\r\n\r\n### T4 — Redirect\r\n\r\n```text\r\nThanks for reaching out! This one is actually better suited for {correct location}.\r\n{brief explanation of why}\r\nFeel free to open it there — they'll be able to help!\r\n```\r\n\r\n### T5 — Acknowledgment\r\n\r\n```text\r\nGood catch, {author}. We've confirmed this is a real issue.\r\n{what we know so far}\r\nWe'll update this thread when we have a fix. Thanks for flagging it!\r\n```\r\n\r\n### T6 — Closing\r\n\r\n```text\r\nThis should be resolved in {version\u002FPR}! 🎉\r\n{brief summary of what changed}\r\nThanks for reporting this, {author} — it made Squad better.\r\n```\r\n\r\n### T7 — Technical Uncertainty\r\n\r\n```text\r\nInteresting find, {author}. We're not 100% sure what's causing this yet.\r\nHere's what we've ruled out: {list}\r\nWe'd love more context if you have it — {specific ask}.\r\nWe'll dig deeper and update this thread.\r\n```\r\n\r\n### T8 — Empathetic Disagreement\r\n\r\n```text\r\nWe hear you, {author}. That's a fair concern.\r\n\r\nThe current design choice was driven by {reason}. We know it's not ideal for every use case.\r\n\r\n{what alternatives exist or what trade-off was made}\r\n\r\nIf you have ideas for how to make this work better for your scenario, we'd love to hear them — open a discussion or drop your thoughts here!\r\n```\r\n\r\n### T9 — Information Request\r\n\r\n```text\r\nThanks for reporting this, {author}!\r\n\r\nTo help us dig into this, could you share:\r\n- {specific ask 1}\r\n- {specific ask 2}\r\n- {specific ask 3, if applicable}\r\n\r\nThat context will help us narrow down what's happening. Appreciate it!\r\n```\r\n\r\n## Anti-Patterns\r\n\r\n- ❌ Posting without human review (NEVER — this is the cardinal rule)\r\n- ❌ Drafting without reading full thread (context is everything)\r\n- ❌ Ignoring confidence flags (🔴 items need Lead or human review)\r\n- ❌ Scanning closed issues (only open items)\r\n- ❌ Responding to issues labeled `squad:internal` or `wontfix`\r\n- ❌ Skipping audit logging (every action must be recorded)\r\n- ❌ Drafting for issues where a squad member already responded (avoid duplicates)\r\n- ❌ Drafting pull request responses in Phase 1 (issues\u002Fdiscussions only)\r\n- ❌ Treating templates like loose examples instead of reusable drafting assets\r\n- ❌ Asking for more info without specific requests\r\n",{"name":5,"tools":50,"domain":67,"source":68,"confidence":69,"description":7},[51,55,59,63],{"name":52,"when":53,"description":54},"github-mcp-server-list_issues","Use for recent open issue scans before thread-level review","List open issues for scan candidates and lightweight triage",{"name":56,"when":57,"description":58},"github-mcp-server-issue_read","Use after selecting a candidate so the communications owner has complete thread context","Read the full issue, comments, and labels before drafting",{"name":60,"when":61,"description":62},"github-mcp-server-search_issues","Use when filtering by keywords, labels, or duplicate response checks","Search for candidate issues or prior squad responses",{"name":64,"when":65,"description":66},"gh CLI","Use gh issue list\u002Fcomment and gh api or gh api graphql when MCP coverage is incomplete","Fallback for GitHub issue comments and discussions workflows","community, communication, workflow","manual (external communications workflow)","low","dev","675c9e97bd5e5e9d3906b175147ddb7eff997499","packages\u002Fsquad-sdk\u002Ftemplates\u002Fskills\u002Fexternal-comms",520,"https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad","https:\u002F\u002Fgithub.github.io\u002Fsquad\u002F","https:\u002F\u002Fgithub.com\u002Fgithub","https:\u002F\u002Fgithub.com\u002Fabout","How people build software.","2026-11-15 02:13:59",{"source":8,"skillName":5,"commands":81,"prompt":91,"zipAvailable":92},[82,85,88],{"tool":83,"command":84},"npx","npx skills add github\u002Fsquad --skill external-comms",{"tool":86,"command":87},"bunx","bunx skills add github\u002Fsquad --skill external-comms",{"tool":89,"command":90},"pnpm","pnpm dlx skills add github\u002Fsquad --skill external-comms","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fpackages\u002Fsquad-sdk\u002Ftemplates\u002Fskills\u002Fexternal-comms\u002FSKILL.md",true,[],[95,103,110,117,124,131],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":102},"01M76752CVNE7K6NWD59CRAYNJ","windows-compatibility","Cross-platform path handling and command patterns","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fplugins\u002Fsquad\u002Fskills\u002Fwindows-compatibility\u002FSKILL.md",[],[],"2026-11-12 13:55:43",{"skillNo":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":102},"01M767508W8A4G83NX1FKEW5VN","versioning-policy","SemVer rules for Squad's stable, preview, insider, and local package versions","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fplugins\u002Fsquad\u002Fskills\u002Fversioning-policy\u002FSKILL.md",[],[],{"skillNo":111,"name":112,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":102},"01M7674X5JJKNJ74B3SS8PGXG2","tiered-memory","Three-tier agent memory model (hot\u002Fcold\u002Fwiki) for context reduction per spawn","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fplugins\u002Fsquad\u002Fskills\u002Ftiered-memory\u002FSKILL.md",[],[],{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":102},"01M7674V3S9H03A76H75M1KSAF","test-discipline","Design assertions that prove the actual contract and fail on realistic regressions","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fplugins\u002Fsquad\u002Fskills\u002Ftest-discipline\u002FSKILL.md",[],[],{"skillNo":125,"name":126,"nameZh":6,"description":127,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":128,"categories":129,"keywords":130,"publishedAt":102},"01M7674QB8CA61RNDX2Z0NAK71","squad","Squad's command catalog and interactive menu. Invoke via \u002Fsquad (slash command) or natural language (\"squad commands\", \"what can squad do\", \"show me squad options\"). Presents categorized operations (Install & Upgrade, Team Management, Issues & PRs, Plugins & Skills, Model & Cost, Sessions & State) as an interactive picker. Routes to the right squad CLI command or the Squad coordinator agent.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fplugins\u002Fsquad\u002Fskills\u002Fsquad\u002FSKILL.md",[],[],{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":137,"publishedAt":102},"01M7674KTQ632N30HQB1TP7TXH","squad-version-check","Internals of how @bradygaster\u002Fsquad-cli stamps its version, how `squad upgrade` works (what it preserves vs overwrites), and how to probe the npm registry for the latest version from a coordinator prompt.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fsquad\u002Fblob\u002Fdev\u002Fplugins\u002Fsquad\u002Fskills\u002Fsquad-version-check\u002FSKILL.md",[],[]]