[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AVED1GD7BAN2RGEHHAND83":3,"skill-same:01M7AVED1GD7BAN2RGEHHAND83":55},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":29,"commitSha":30,"skillDirectory":5,"forksCount":14,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M7AVED1GD7BAN2RGEHHAND83","provenance-tracing",null,"Find where an IP\u002Fclaim came from; verify what you said.","peterlau123\u002Fskills-you-need","peterlau123","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F19188200?v=4","01M7AV0Y55C30Y4E2TGRQ8X7WR","Sloth","01M7AV0Y5SNAJQ4E3FG1EEK8S4",0,"MIT","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fprovenance-tracing\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: provenance-tracing\ndescription: Find where an IP\u002Fclaim came from; verify what you said.\nversion: 1.0.0\nauthor: Hermes Agent\nlicense: MIT\nplatforms: [linux, macos]\n---\n\n# Provenance Tracing (where did X come from?)\n\nUse when the user asks \"where did this IP\u002Fhostname\u002Fclaim come from?\", \"did you ever say X?\", \"who told you Y?\", or challenges a fact attributed to you. Goal: answer with evidence, attribute the origin to a specific source (user message \u002F assistant output \u002F tool output \u002F file \u002F config), and never guess or fabricate.\n\n## Golden rule\nNever answer from memory or assumption — prove it from primary sources. The user's question may itself be based on a misremembering; verify what YOU actually said before accepting or denying anything.\n\n## Step 1 — FTS5 session search (fast, but NOT exhaustive)\n`session_search(query=...)` is a fine first pass, but has two blind spots:\n- The FTS5 unicode61 tokenizer splits on punctuation (dots, slashes, colons). IP addresses (`10.10.192.55`), FQDNs, and version strings (`3.9.10`) do NOT match as phrase queries even when present in the DB.\n- The default `role_filter='user,assistant'` excludes tool output.\n\n→ **Zero results from session_search is NOT proof of absence.** Always confirm with Step 2.\n\n## Step 2 — Raw DB search (definitive ground truth)\nLocate the session DB (usually `~\u002F.hermes\u002Fprofiles\u002F\u003Cprofile>\u002Fstate.db`) and search it raw:\n\n```bash\nfind ~\u002F.hermes \\( -name \"*.db\" -o -name \"*.sqlite*\" \\) 2>\u002Fdev\u002Fnull\nstrings \u003Cstate.db> | grep -c \"\u003Cpattern>\"   # 0 = truly absent; >0 = present somewhere\n```\n\nThen pin down the exact rows with Python sqlite3 (plain `LIKE`, NOT the FTS table):\n\n```sql\nSELECT m.id, m.session_id, m.role, m.tool_name, m.timestamp, substr(m.content,1,3000)\nFROM messages m\nWHERE m.content LIKE '%\u003Cpattern>%' OR m.tool_calls LIKE '%\u003Cpattern>%' OR m.reasoning LIKE '%\u003Cpattern>%'\nORDER BY m.timestamp;\n```\n\nAlso scan `sessions` for title\u002Fsource. **Crucial attribution step**: check `role` — if the only hits are `role='user'` messages, or assistant tool calls that merely echoed the user's text, then the origin is the user's own message, not something you said.\n\n## Step 3 — Filesystem & config sweep\nGrep the string in: the repo in question, `\u002Fetc\u002Fhosts`, `~\u002F.ssh\u002Fconfig`, `~\u002F.gitconfig`, env vars, `~\u002F.hermes` configs.\nCAUTION: `agent.log` \u002F `gateway.log` echo inbound user messages — a hit in logs may just be the user's own text quoted back. Same for your own tool-output echo. Neither is evidence you originated it.\n\n## Step 4 — Report with an attribution table\nPresent: scope checked | result (e.g. session DB, repo files, \u002Fetc\u002Fhosts, ssh config, git config, env, logs). State plainly who originated it (user vs assistant vs file). If the user is wrong (\"you never said that\", \"it's in no config\"), say so with the evidence and offer the likely real origin (another machine, another environment, another conversation) — don't let a false accusation stand, and don't invent a source to be agreeable.\n\n## Pitfalls\n- **FTS5 zero results ≠ absent** — Step 2 is the ground truth (see references\u002Fip-address-case.md).\n- A string found in your own tool outputs that quoted the user's message is a false positive for \"assistant originated it\".\n- Don't stop at \"the string exists in the DB\" — identify the role of the message that FIRST introduced it.\n- When the user asserts a capability difference (\"direct access works\", \"that URL is reachable\"), verify live with a probe rather than accepting or repeating stale claims — see `github-restricted-network-access` for the probe ladder.\n\n## References\n- `references\u002Fip-address-case.md` — worked example: \"where did 10.10.192.55 come from?\" (answer: the user's own message; FTS5 missed it entirely).\n","# Provenance Tracing (where did X come from?)\n\nUse when the user asks \"where did this IP\u002Fhostname\u002Fclaim come from?\", \"did you ever say X?\", \"who told you Y?\", or challenges a fact attributed to you. Goal: answer with evidence, attribute the origin to a specific source (user message \u002F assistant output \u002F tool output \u002F file \u002F config), and never guess or fabricate.\n\n## Golden rule\nNever answer from memory or assumption — prove it from primary sources. The user's question may itself be based on a misremembering; verify what YOU actually said before accepting or denying anything.\n\n## Step 1 — FTS5 session search (fast, but NOT exhaustive)\n`session_search(query=...)` is a fine first pass, but has two blind spots:\n- The FTS5 unicode61 tokenizer splits on punctuation (dots, slashes, colons). IP addresses (`10.10.192.55`), FQDNs, and version strings (`3.9.10`) do NOT match as phrase queries even when present in the DB.\n- The default `role_filter='user,assistant'` excludes tool output.\n\n→ **Zero results from session_search is NOT proof of absence.** Always confirm with Step 2.\n\n## Step 2 — Raw DB search (definitive ground truth)\nLocate the session DB (usually `~\u002F.hermes\u002Fprofiles\u002F\u003Cprofile>\u002Fstate.db`) and search it raw:\n\n```bash\nfind ~\u002F.hermes \\( -name \"*.db\" -o -name \"*.sqlite*\" \\) 2>\u002Fdev\u002Fnull\nstrings \u003Cstate.db> | grep -c \"\u003Cpattern>\"   # 0 = truly absent; >0 = present somewhere\n```\n\nThen pin down the exact rows with Python sqlite3 (plain `LIKE`, NOT the FTS table):\n\n```sql\nSELECT m.id, m.session_id, m.role, m.tool_name, m.timestamp, substr(m.content,1,3000)\nFROM messages m\nWHERE m.content LIKE '%\u003Cpattern>%' OR m.tool_calls LIKE '%\u003Cpattern>%' OR m.reasoning LIKE '%\u003Cpattern>%'\nORDER BY m.timestamp;\n```\n\nAlso scan `sessions` for title\u002Fsource. **Crucial attribution step**: check `role` — if the only hits are `role='user'` messages, or assistant tool calls that merely echoed the user's text, then the origin is the user's own message, not something you said.\n\n## Step 3 — Filesystem & config sweep\nGrep the string in: the repo in question, `\u002Fetc\u002Fhosts`, `~\u002F.ssh\u002Fconfig`, `~\u002F.gitconfig`, env vars, `~\u002F.hermes` configs.\nCAUTION: `agent.log` \u002F `gateway.log` echo inbound user messages — a hit in logs may just be the user's own text quoted back. Same for your own tool-output echo. Neither is evidence you originated it.\n\n## Step 4 — Report with an attribution table\nPresent: scope checked | result (e.g. session DB, repo files, \u002Fetc\u002Fhosts, ssh config, git config, env, logs). State plainly who originated it (user vs assistant vs file). If the user is wrong (\"you never said that\", \"it's in no config\"), say so with the evidence and offer the likely real origin (another machine, another environment, another conversation) — don't let a false accusation stand, and don't invent a source to be agreeable.\n\n## Pitfalls\n- **FTS5 zero results ≠ absent** — Step 2 is the ground truth (see references\u002Fip-address-case.md).\n- A string found in your own tool outputs that quoted the user's message is a false positive for \"assistant originated it\".\n- Don't stop at \"the string exists in the DB\" — identify the role of the message that FIRST introduced it.\n- When the user asserts a capability difference (\"direct access works\", \"that URL is reachable\"), verify live with a probe rather than accepting or repeating stale claims — see `github-restricted-network-access` for the probe ladder.\n\n## References\n- `references\u002Fip-address-case.md` — worked example: \"where did 10.10.192.55 come from?\" (answer: the user's own message; FTS5 missed it entirely).\n",{"name":5,"author":24,"license":15,"version":25,"platforms":26,"description":7},"Hermes Agent","1.0.0",[27,28],"linux","macos","main","c0b220603dea88e986e5b31268098ef2f5b07fd0","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need","https:\u002F\u002Fgithub.com\u002Fpeterlau123","A software engineer in deep learning\r\n","2026-08-19 16:46:00",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add peterlau123\u002Fskills-you-need --skill provenance-tracing",{"tool":41,"command":42},"bunx","bunx skills add peterlau123\u002Fskills-you-need --skill provenance-tracing",{"tool":44,"command":45},"pnpm","pnpm dlx skills add peterlau123\u002Fskills-you-need --skill provenance-tracing","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fprovenance-tracing\u002FSKILL.md",true,[49],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fip-address-case.md","REFERENCE","参考文档",2673,"https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fprovenance-tracing\u002Freferences\u002Fip-address-case.md",[56,63,70,77,84,91],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7AVH2D16QBRC8T2P6SVXSTS","jekyll-chirpy-blog","Use when editing the Jekyll\u002FChirpy blog or fixing mermaid.","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fweb\u002Fjekyll-chirpy-blog\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7AVGYYFDK4GKCSNHBCP59Y6","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fsoftware-development\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7AVGT9GAVB5EBKZ2616R27Z","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fsoftware-development\u002Fsystematic-debugging\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7AVGQ23T5EHRQP5GZ45KZC5","spike","Throwaway experiments to validate an idea before build.","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fsoftware-development\u002Fspike\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7AVGKZ0ZD91ATKD9QB44E1B","simplify-code","Parallel 4-agent cleanup of recent code changes.","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fsoftware-development\u002Fsimplify-code\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":19},"01M7AVGFBR76BWQNG2X4CN5SQT","requesting-code-review","Pre-commit review: security scan, quality gates, auto-fix.","https:\u002F\u002Fgithub.com\u002Fpeterlau123\u002Fskills-you-need\u002Fblob\u002Fmain\u002Fsoftware-development\u002Frequesting-code-review\u002FSKILL.md",[],[]]