[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CZ9TV1M37BH9FWKQ003RVB":3,"skill-same:01M6CZ9TV1M37BH9FWKQ003RVB":74},{"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":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":56,"repositoryHomepageUrl":57,"ownerHtmlUrl":58,"ownerBlogUrl":57,"ownerDescription":59,"updatedAt":60,"install":61,"resources":73},"01M6CZ9TV1M37BH9FWKQ003RVB","narrative-tracker",null,"Track rising, peaking, and fading crypto\u002Ftech narratives with quantitative mindshare + velocity signals and explicit positioning calls","aeonfun\u002Faeon","aeonfun","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F287715280?v=4","01M6CZ23C6MXGMF4N0V1DZ1247","aeon","01M6CZ23GMZP1BR424J8Y8K3DG",793,"MIT","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fskills\u002Fnarrative-tracker\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"tracker","narrative","narrative tracker","signals","track","true","calls","read-only","mindshare","with","tech","research","positioning","velocity","crypto","only","write","read","xai","api","2026-11-10 10:12:30",false,"---\nname: narrative-tracker\ndescription: Track rising, peaking, and fading crypto\u002Ftech narratives with quantitative mindshare + velocity signals and explicit positioning calls\nmetadata:\n  commits: true\n  permissions:\n    - contents:write\n  title: Narrative Tracker\n  mode: read-only\n  category: crypto\n  tags:\n    - crypto\n    - research\n  requires:\n    - XAI_API_KEY\n---\n\u003C!-- autoresearch: variation B — sharper output (quantitative mindshare + velocity + explicit positioning calls, with multi-angle inputs from A, dedup\u002Fempty-state handling from C, and transition detection from D) -->\n\nRead `memory\u002FMEMORY.md` for context on prior narrative observations.\nRead the last 3 days of `memory\u002Flogs\u002F` — specifically any prior `### narrative-tracker` entries — to (a) avoid re-reporting the same narratives without new info, and (b) detect phase transitions vs the last run.\n\n## Goal\n\nProduce a *decision-grade* narrative map: every narrative gets a mindshare score, a velocity arrow, a sentiment tag, named drivers, and an explicit position call. Classification without a position call is noise.\n\n## Steps\n\n### 1. Ingest signals\n\n**a. X\u002FTwitter narratives - X.AI API (primary).** The primary signal is a direct `curl` to the X.AI Responses API with Grok's `x_search` tool - see **## Fetching** below for the full contract. `XAI_API_KEY` is injected into this skill's environment via `requires:`; it is present and valid, so this path is required whenever the key check gives `KEY_PRESENT`. Set the Bash tool `timeout` to ≥180000 when running the curl (x_search takes 30-120s) and capture the HTTP status.\n\n**Shell shape.** Each Bash call is one allowed command (pipes are fine). The run refuses a call that sets a variable (`FROM_DATE=$(date ...)`, `PROMPT=\"...\"`, `HTTP=$(.\u002Fsecretcurl ...)`), uses `[ ... ]` tests, or joins commands with `;`, `&&` or `||`, and each refusal costs a model turn. So run each block below as its own Bash call, in order: get a date with a bare `date` call and write the printed value into later commands (the `\u003CFROM_DATE>` \u002F `\u003CTO_DATE>` placeholders); put the prompt and dates inline in the `jq -n` filter (a jq `$name` variable such as `--arg p ... '$p'` is refused too) and pipe it straight into `.\u002Fsecretcurl`; never put a `$VAR` in the `.\u002Fsecretcurl` line.\n\nGet the window dates, each `date` as its own call:\n```bash\ndate -u -d '3 days ago' +%Y-%m-%d\n```\n```bash\ndate -u +%Y-%m-%d\n```\nThe first prints FROM_DATE, the second TO_DATE.\n\nKey check: read the \"Keys set for this run\" line in your run notes, never a shell test (those are refused and cost a turn). `XAI_API_KEY` on that line = `KEY_PRESENT`, not on it = `KEY_UNSET`.\n\nBuild the JSON payload with jq (do NOT hand-assemble it), with the dates written in literally, and pipe it straight into the POST in the same call. Do not write it to a file first: a `>` redirect into a file is refused in BOTH tiers, \u002Ftmp included, and this read-only run has no Write tool. `.\u002Fsecretcurl` reads the body from stdin (`--data-binary @-`; its own config goes over a separate pipe, so stdin is free):\n```bash\njq -n '{model:\"grok-4.7\", input:[{role:\"user\",content:\"Search X for the dominant crypto and tech narratives from \u003CFROM_DATE> to \u003CTO_DATE>. Return 12-15 distinct narrative threads. For each: 1) short label, 2) 3-5 representative @handles driving it, 3) 2-3 tweet permalinks, 4) rough mention-volume descriptor (niche \u002F growing \u002F saturating \u002F cooling), 5) the strongest one-line bear case against it.\"}], tools:[{type:\"x_search\",from_date:\"\u003CFROM_DATE>\",to_date:\"\u003CTO_DATE>\"}]}' \\\n  | .\u002Fsecretcurl -s -o \u002Ftmp\u002Fxai-nt.json -w '%{http_code}' --max-time 150 -X POST \"https:\u002F\u002Fapi.x.ai\u002Fv1\u002Fresponses\" \\\n  -H \"Content-Type: application\u002Fjson\" \\\n  -H \"Authorization: Bearer {XAI_API_KEY}\" \\\n  --data-binary @-\n```\nRun that block **verbatim** apart from the placeholders (do not hand-reassemble the JSON - the `jq -n` builder exists precisely so quoting\u002Fexpansion can't break). The curl prints the status code (`HTTP`), and `.\u002Fsecretcurl` prints a `secretcurl xai attempt=... http=\u003Ccode> bytes=\u003Cn> ...` line on stderr; that `xai ... http=` line **must appear in your output** - it is your proof the call ran. On `HTTP=200` with a non-empty body, parse `\u002Ftmp\u002Fxai-nt.json` with `jq -r '.output[] | select(.type == \"message\") | .content[] | select(.type == \"output_text\") | .text'` and use that as the primary narrative signal (`SOURCE=api`).\n\n**b. WebSearch \u002F WebFetch fallback (last-resort only).** You may reach for this **only after** you have shown a `secretcurl xai ... http=\u003Ccode>` line proving Path A actually ran and returned a non-2xx code (or an empty body, or timed out). If you have no `xai ... http=` line, you did not run the call - go back and run it. Reach for the fallback **only** when Path A genuinely fails - `KEY_UNSET`, a non-2xx HTTP code, an empty parse, or a timeout. It is lower quality (WebSearch favours old high-engagement posts) and is **never co-equal** with Path A. Record the fetch failure in your final output with the **true reason** - `key-unset` | `http-\u003Ccode>` | `empty` | `timeout` - never \"XAI_API_KEY unavailable\" when the key was set (a slow curl is a `timeout`, not a missing key). Then compile narratives via WebSearch (`crypto narrative ${TO_DATE}`, `AI agent crypto trend this week`) and WebFetch on individual tweet URLs; discard anything older than the 3-day window.\n\n**c. Quantitative reference points (supplement).** Independently of the fetch path, cross-check mindshare against external quantitative benchmarks with one WebSearch: `DefiLlama narrative tracker` OR `Kaito mindshare leaderboard`. Pull 1-2 concrete numbers (project name, metric, link) to calibrate the mindshare scores in step 2. This is a calibration cross-check, **not** a narrative source. Do not paraphrase — extract facts.\n\n**d. Memory diff.** Extract narrative labels mentioned in the last 3 days of `### narrative-tracker` log entries. You'll compare against them in step 4.\n\n### 2. Score each narrative\n\nFor each distinct narrative (merge near-duplicates aggressively — \"AI agents\" and \"agentic crypto\" are the same), assign:\n\n| Field | Scale | How to decide |\n|---|---|---|\n| **Mindshare** | 1-5 | 1 = fringe, 3 = known in the sector, 5 = dominating timelines. Base on count of distinct drivers + whether you had to dig or it surfaced unprompted. |\n| **Velocity** | ↑↑ \u002F ↑ \u002F → \u002F ↓ \u002F ↓↓ | Compared to the 3-day window or prior log entries. ↑↑ = tripled in attention, ↓↓ = was loud 3 days ago, now absent. |\n| **Phase** | Emerging \u002F Rising \u002F Peak \u002F Fading | Use the velocity + mindshare combo. Emerging = low mindshare, high velocity. Peak = high mindshare, flat\u002Fdown velocity. Fading = high mindshare last week, now ↓. |\n| **Sentiment** | Bull \u002F Mixed \u002F Bear \u002F Cope | Cope = bag-holder energy, bear narratives dressed as bull takes. |\n| **Drivers** | 2-3 named | Accounts, projects, or funds amplifying it. Include @handles. |\n| **Bear case** | 1 line | The sharpest argument against. If the consensus is obviously right, say so and mark \"no contrarian edge\". |\n| **Position** | FRONT-RUN \u002F RIDE \u002F FADE \u002F WATCH \u002F IGNORE | FRONT-RUN = emerging + contrarian edge. RIDE = rising, not yet peaked. FADE = peak with weak fundamentals or reflexivity flip. WATCH = unclear. IGNORE = mindshare 1-2 with no catalyst. |\n\nDrop any narrative that ends up IGNORE unless it's structurally important — noise reduction is the goal.\n\n### 3. Detect transitions\n\nCompare today's narratives to the last 3 days of logs:\n- **NEW** — narrative wasn't in prior logs at all\n- **PROMOTED** — phase moved up (e.g. Emerging → Rising)\n- **DEMOTED** — phase moved down\n- **DEAD** — was in prior logs, now absent from all signals\n\nTransitions are the highest-value output — the point of a daily tracker is to catch inflection points, not re-report the zeitgeist.\n\n### 4. Flag reflexivity\n\nFor each narrative, flag if the story itself is moving outcomes:\n- Token prices moving on narrative alone (no fundamentals shift)\n- Projects rebranding\u002Fpivoting to ride the narrative\n- VCs publicly endorsing to manufacture legitimacy\n- Prediction markets or on-chain flows reflecting narrative belief\n\nOnly flag explicit cases with a concrete example. \"Reflexivity\" without evidence is hand-waving.\n\n### 5. Format the notification\n\nKeep under 4000 chars. Lead with transitions and reflexivity — those are the decisions. Classification goes below.\n\n```\n*Narrative Tracker - ${today}*\n\nTRANSITIONS\n• NEW: \u003Clabel> - \u003Cwhy it matters> - \u003Clink>\n• PROMOTED: \u003Clabel> Rising → Peak - \u003Cwhat flipped>\n• DEMOTED: \u003Clabel> Peak → Fading - \u003Cwhat cooled>\n• DEAD: \u003Clabel> - gone\n\nREFLEXIVITY ALERT\n• \u003Cnarrative> - \u003Cconcrete evidence the story is moving outcomes>\n\nPOSITIONS\n• FRONT-RUN: \u003Clabel> (mindshare 2 ↑↑, Bull) - \u003Cdriver> - \u003Cbear case> - \u003Clink>\n• RIDE: \u003Clabel> (3 ↑, Bull) - \u003Cdriver> - \u003Cbear case>\n• FADE: \u003Clabel> (5 → Cope) - \u003Cdriver> - \u003Creflexivity note>\n\nMAP\nEmerging: \u003Clabels>\nRising: \u003Clabels>\nPeak: \u003Clabels>\nFading: \u003Clabels>\n```\n\nIf absolutely nothing new or notable (no transitions, no reflexivity, no FRONT-RUN\u002FFADE calls): send a one-line update instead of the full template — `*Narrative Tracker — ${today}*: no phase transitions, map unchanged from \u003Clast_date>.`\n\n### 6. Send via `.\u002Fnotify`\n\n### 7. Log record\n\nThis skill is `read-only`, so the workflow's read-only guard writes its `### narrative-tracker` log entry from your captured output; a self-written entry would be a duplicate. Don't append to `memory\u002Flogs\u002F` yourself - put this record in your **final output**: the full structured output (not just the notification - include all narratives considered, even IGNOREd ones, so future diffs work). If a full run produced nothing actionable, record `NARRATIVE_TRACKER_OK` with the narrative labels seen (so tomorrow's diff still has a baseline).\n\n## Guidelines\n\n- Quantitative over vibes. Every narrative gets mindshare 1-5 and a velocity arrow — no exceptions. If you can't score it, drop it.\n- Transitions > classification. A daily tracker's value is catching moves, not listing the weather.\n- Named drivers only. \"Crypto Twitter is excited about X\" is not a driver. \"@handle + @handle + @fund\" is.\n- Position calls are mandatory for Emerging\u002FRising\u002FPeak narratives. If signals are genuinely ambiguous or contradictory, **WATCH** is an acceptable call — but never omit a position entirely and never invent conviction you don't have.\n- Ruthless dedup. Same narrative under two labels = one narrative. Merge, don't split.\n- Call out cope. Manufactured narratives, coordinated shilling, and dead-cat bounces get tagged explicitly.\n- Prioritize topics tracked in MEMORY.md over generic market chatter.\n\n## Fetching\n\n`XAI_API_KEY` is **injected into this skill's environment** (declared in `requires:`). It is present and valid. **The primary fetch path is a direct `curl` to `https:\u002F\u002Fapi.x.ai\u002Fv1\u002Fresponses` with `Authorization: Bearer {XAI_API_KEY}`** (model `grok-4.7`, `\"tools\":[{\"type\":\"x_search\"}]`). There is **no network sandbox** blocking this - earlier versions of this skill claimed there was, and that is stale and false. Just make the call.\n\nRules:\n1. **Check, don't assume.** Read the \"Keys set for this run\" line in your run notes, never a shell test (those are refused and cost a turn): `XAI_API_KEY` on that line = `KEY_PRESENT`, not on it = `KEY_UNSET`. If `KEY_PRESENT` (it will be), Path A (the curl in step 1a) is required before any fallback.\n2. **Allow enough time.** The `x_search` call typically takes 30-120s. When you invoke the Bash tool for the curl, **set the tool's `timeout` to at least 180000 (180s)** and keep **`--max-time 150`** on the curl so it fails cleanly rather than hanging. A slow curl is **not** a missing key — never treat a timeout as key-unavailable.\n3. **Capture the HTTP status** so the fallback decision is fact-based (see the skeleton in step 1a). `HTTP=200` with a non-empty parsed body → use it.\n4. **Fall back only on a real failure**, recording the true reason — `key-unset` (only if the key check gave `KEY_UNSET`), `http-\u003Ccode>` (non-2xx), `empty` (200 but nothing parsed), or `timeout`. Never write \"XAI_API_KEY unavailable\" when the key was set.\n\n**WebSearch \u002F WebFetch are last-resort fallbacks only** — lower quality, never primary or co-equal. Do not reach for them while the key works.\n\n## Environment Variables Required\n\n- `XAI_API_KEY` — X.AI API key for Grok's `x_search` tool. Declared in `requires:`, so it is **injected into this skill's environment** and is the **primary fetch path** for the narrative signal (step 1a). If it is ever unset, the skill degrades to WebSearch\u002FWebFetch at lower quality.\n- Notification channels configured via repo secrets (see CLAUDE.md).\n","\u003C!-- autoresearch: variation B — sharper output (quantitative mindshare + velocity + explicit positioning calls, with multi-angle inputs from A, dedup\u002Fempty-state handling from C, and transition detection from D) -->\n\nRead `memory\u002FMEMORY.md` for context on prior narrative observations.\nRead the last 3 days of `memory\u002Flogs\u002F` — specifically any prior `### narrative-tracker` entries — to (a) avoid re-reporting the same narratives without new info, and (b) detect phase transitions vs the last run.\n\n## Goal\n\nProduce a *decision-grade* narrative map: every narrative gets a mindshare score, a velocity arrow, a sentiment tag, named drivers, and an explicit position call. Classification without a position call is noise.\n\n## Steps\n\n### 1. Ingest signals\n\n**a. X\u002FTwitter narratives - X.AI API (primary).** The primary signal is a direct `curl` to the X.AI Responses API with Grok's `x_search` tool - see **## Fetching** below for the full contract. `XAI_API_KEY` is injected into this skill's environment via `requires:`; it is present and valid, so this path is required whenever the key check gives `KEY_PRESENT`. Set the Bash tool `timeout` to ≥180000 when running the curl (x_search takes 30-120s) and capture the HTTP status.\n\n**Shell shape.** Each Bash call is one allowed command (pipes are fine). The run refuses a call that sets a variable (`FROM_DATE=$(date ...)`, `PROMPT=\"...\"`, `HTTP=$(.\u002Fsecretcurl ...)`), uses `[ ... ]` tests, or joins commands with `;`, `&&` or `||`, and each refusal costs a model turn. So run each block below as its own Bash call, in order: get a date with a bare `date` call and write the printed value into later commands (the `\u003CFROM_DATE>` \u002F `\u003CTO_DATE>` placeholders); put the prompt and dates inline in the `jq -n` filter (a jq `$name` variable such as `--arg p ... '$p'` is refused too) and pipe it straight into `.\u002Fsecretcurl`; never put a `$VAR` in the `.\u002Fsecretcurl` line.\n\nGet the window dates, each `date` as its own call:\n```bash\ndate -u -d '3 days ago' +%Y-%m-%d\n```\n```bash\ndate -u +%Y-%m-%d\n```\nThe first prints FROM_DATE, the second TO_DATE.\n\nKey check: read the \"Keys set for this run\" line in your run notes, never a shell test (those are refused and cost a turn). `XAI_API_KEY` on that line = `KEY_PRESENT`, not on it = `KEY_UNSET`.\n\nBuild the JSON payload with jq (do NOT hand-assemble it), with the dates written in literally, and pipe it straight into the POST in the same call. Do not write it to a file first: a `>` redirect into a file is refused in BOTH tiers, \u002Ftmp included, and this read-only run has no Write tool. `.\u002Fsecretcurl` reads the body from stdin (`--data-binary @-`; its own config goes over a separate pipe, so stdin is free):\n```bash\njq -n '{model:\"grok-4.7\", input:[{role:\"user\",content:\"Search X for the dominant crypto and tech narratives from \u003CFROM_DATE> to \u003CTO_DATE>. Return 12-15 distinct narrative threads. For each: 1) short label, 2) 3-5 representative @handles driving it, 3) 2-3 tweet permalinks, 4) rough mention-volume descriptor (niche \u002F growing \u002F saturating \u002F cooling), 5) the strongest one-line bear case against it.\"}], tools:[{type:\"x_search\",from_date:\"\u003CFROM_DATE>\",to_date:\"\u003CTO_DATE>\"}]}' \\\n  | .\u002Fsecretcurl -s -o \u002Ftmp\u002Fxai-nt.json -w '%{http_code}' --max-time 150 -X POST \"https:\u002F\u002Fapi.x.ai\u002Fv1\u002Fresponses\" \\\n  -H \"Content-Type: application\u002Fjson\" \\\n  -H \"Authorization: Bearer {XAI_API_KEY}\" \\\n  --data-binary @-\n```\nRun that block **verbatim** apart from the placeholders (do not hand-reassemble the JSON - the `jq -n` builder exists precisely so quoting\u002Fexpansion can't break). The curl prints the status code (`HTTP`), and `.\u002Fsecretcurl` prints a `secretcurl xai attempt=... http=\u003Ccode> bytes=\u003Cn> ...` line on stderr; that `xai ... http=` line **must appear in your output** - it is your proof the call ran. On `HTTP=200` with a non-empty body, parse `\u002Ftmp\u002Fxai-nt.json` with `jq -r '.output[] | select(.type == \"message\") | .content[] | select(.type == \"output_text\") | .text'` and use that as the primary narrative signal (`SOURCE=api`).\n\n**b. WebSearch \u002F WebFetch fallback (last-resort only).** You may reach for this **only after** you have shown a `secretcurl xai ... http=\u003Ccode>` line proving Path A actually ran and returned a non-2xx code (or an empty body, or timed out). If you have no `xai ... http=` line, you did not run the call - go back and run it. Reach for the fallback **only** when Path A genuinely fails - `KEY_UNSET`, a non-2xx HTTP code, an empty parse, or a timeout. It is lower quality (WebSearch favours old high-engagement posts) and is **never co-equal** with Path A. Record the fetch failure in your final output with the **true reason** - `key-unset` | `http-\u003Ccode>` | `empty` | `timeout` - never \"XAI_API_KEY unavailable\" when the key was set (a slow curl is a `timeout`, not a missing key). Then compile narratives via WebSearch (`crypto narrative ${TO_DATE}`, `AI agent crypto trend this week`) and WebFetch on individual tweet URLs; discard anything older than the 3-day window.\n\n**c. Quantitative reference points (supplement).** Independently of the fetch path, cross-check mindshare against external quantitative benchmarks with one WebSearch: `DefiLlama narrative tracker` OR `Kaito mindshare leaderboard`. Pull 1-2 concrete numbers (project name, metric, link) to calibrate the mindshare scores in step 2. This is a calibration cross-check, **not** a narrative source. Do not paraphrase — extract facts.\n\n**d. Memory diff.** Extract narrative labels mentioned in the last 3 days of `### narrative-tracker` log entries. You'll compare against them in step 4.\n\n### 2. Score each narrative\n\nFor each distinct narrative (merge near-duplicates aggressively — \"AI agents\" and \"agentic crypto\" are the same), assign:\n\n| Field | Scale | How to decide |\n|---|---|---|\n| **Mindshare** | 1-5 | 1 = fringe, 3 = known in the sector, 5 = dominating timelines. Base on count of distinct drivers + whether you had to dig or it surfaced unprompted. |\n| **Velocity** | ↑↑ \u002F ↑ \u002F → \u002F ↓ \u002F ↓↓ | Compared to the 3-day window or prior log entries. ↑↑ = tripled in attention, ↓↓ = was loud 3 days ago, now absent. |\n| **Phase** | Emerging \u002F Rising \u002F Peak \u002F Fading | Use the velocity + mindshare combo. Emerging = low mindshare, high velocity. Peak = high mindshare, flat\u002Fdown velocity. Fading = high mindshare last week, now ↓. |\n| **Sentiment** | Bull \u002F Mixed \u002F Bear \u002F Cope | Cope = bag-holder energy, bear narratives dressed as bull takes. |\n| **Drivers** | 2-3 named | Accounts, projects, or funds amplifying it. Include @handles. |\n| **Bear case** | 1 line | The sharpest argument against. If the consensus is obviously right, say so and mark \"no contrarian edge\". |\n| **Position** | FRONT-RUN \u002F RIDE \u002F FADE \u002F WATCH \u002F IGNORE | FRONT-RUN = emerging + contrarian edge. RIDE = rising, not yet peaked. FADE = peak with weak fundamentals or reflexivity flip. WATCH = unclear. IGNORE = mindshare 1-2 with no catalyst. |\n\nDrop any narrative that ends up IGNORE unless it's structurally important — noise reduction is the goal.\n\n### 3. Detect transitions\n\nCompare today's narratives to the last 3 days of logs:\n- **NEW** — narrative wasn't in prior logs at all\n- **PROMOTED** — phase moved up (e.g. Emerging → Rising)\n- **DEMOTED** — phase moved down\n- **DEAD** — was in prior logs, now absent from all signals\n\nTransitions are the highest-value output — the point of a daily tracker is to catch inflection points, not re-report the zeitgeist.\n\n### 4. Flag reflexivity\n\nFor each narrative, flag if the story itself is moving outcomes:\n- Token prices moving on narrative alone (no fundamentals shift)\n- Projects rebranding\u002Fpivoting to ride the narrative\n- VCs publicly endorsing to manufacture legitimacy\n- Prediction markets or on-chain flows reflecting narrative belief\n\nOnly flag explicit cases with a concrete example. \"Reflexivity\" without evidence is hand-waving.\n\n### 5. Format the notification\n\nKeep under 4000 chars. Lead with transitions and reflexivity — those are the decisions. Classification goes below.\n\n```\n*Narrative Tracker - ${today}*\n\nTRANSITIONS\n• NEW: \u003Clabel> - \u003Cwhy it matters> - \u003Clink>\n• PROMOTED: \u003Clabel> Rising → Peak - \u003Cwhat flipped>\n• DEMOTED: \u003Clabel> Peak → Fading - \u003Cwhat cooled>\n• DEAD: \u003Clabel> - gone\n\nREFLEXIVITY ALERT\n• \u003Cnarrative> - \u003Cconcrete evidence the story is moving outcomes>\n\nPOSITIONS\n• FRONT-RUN: \u003Clabel> (mindshare 2 ↑↑, Bull) - \u003Cdriver> - \u003Cbear case> - \u003Clink>\n• RIDE: \u003Clabel> (3 ↑, Bull) - \u003Cdriver> - \u003Cbear case>\n• FADE: \u003Clabel> (5 → Cope) - \u003Cdriver> - \u003Creflexivity note>\n\nMAP\nEmerging: \u003Clabels>\nRising: \u003Clabels>\nPeak: \u003Clabels>\nFading: \u003Clabels>\n```\n\nIf absolutely nothing new or notable (no transitions, no reflexivity, no FRONT-RUN\u002FFADE calls): send a one-line update instead of the full template — `*Narrative Tracker — ${today}*: no phase transitions, map unchanged from \u003Clast_date>.`\n\n### 6. Send via `.\u002Fnotify`\n\n### 7. Log record\n\nThis skill is `read-only`, so the workflow's read-only guard writes its `### narrative-tracker` log entry from your captured output; a self-written entry would be a duplicate. Don't append to `memory\u002Flogs\u002F` yourself - put this record in your **final output**: the full structured output (not just the notification - include all narratives considered, even IGNOREd ones, so future diffs work). If a full run produced nothing actionable, record `NARRATIVE_TRACKER_OK` with the narrative labels seen (so tomorrow's diff still has a baseline).\n\n## Guidelines\n\n- Quantitative over vibes. Every narrative gets mindshare 1-5 and a velocity arrow — no exceptions. If you can't score it, drop it.\n- Transitions > classification. A daily tracker's value is catching moves, not listing the weather.\n- Named drivers only. \"Crypto Twitter is excited about X\" is not a driver. \"@handle + @handle + @fund\" is.\n- Position calls are mandatory for Emerging\u002FRising\u002FPeak narratives. If signals are genuinely ambiguous or contradictory, **WATCH** is an acceptable call — but never omit a position entirely and never invent conviction you don't have.\n- Ruthless dedup. Same narrative under two labels = one narrative. Merge, don't split.\n- Call out cope. Manufactured narratives, coordinated shilling, and dead-cat bounces get tagged explicitly.\n- Prioritize topics tracked in MEMORY.md over generic market chatter.\n\n## Fetching\n\n`XAI_API_KEY` is **injected into this skill's environment** (declared in `requires:`). It is present and valid. **The primary fetch path is a direct `curl` to `https:\u002F\u002Fapi.x.ai\u002Fv1\u002Fresponses` with `Authorization: Bearer {XAI_API_KEY}`** (model `grok-4.7`, `\"tools\":[{\"type\":\"x_search\"}]`). There is **no network sandbox** blocking this - earlier versions of this skill claimed there was, and that is stale and false. Just make the call.\n\nRules:\n1. **Check, don't assume.** Read the \"Keys set for this run\" line in your run notes, never a shell test (those are refused and cost a turn): `XAI_API_KEY` on that line = `KEY_PRESENT`, not on it = `KEY_UNSET`. If `KEY_PRESENT` (it will be), Path A (the curl in step 1a) is required before any fallback.\n2. **Allow enough time.** The `x_search` call typically takes 30-120s. When you invoke the Bash tool for the curl, **set the tool's `timeout` to at least 180000 (180s)** and keep **`--max-time 150`** on the curl so it fails cleanly rather than hanging. A slow curl is **not** a missing key — never treat a timeout as key-unavailable.\n3. **Capture the HTTP status** so the fallback decision is fact-based (see the skeleton in step 1a). `HTTP=200` with a non-empty parsed body → use it.\n4. **Fall back only on a real failure**, recording the true reason — `key-unset` (only if the key check gave `KEY_UNSET`), `http-\u003Ccode>` (non-2xx), `empty` (200 but nothing parsed), or `timeout`. Never write \"XAI_API_KEY unavailable\" when the key was set.\n\n**WebSearch \u002F WebFetch are last-resort fallbacks only** — lower quality, never primary or co-equal. Do not reach for them while the key works.\n\n## Environment Variables Required\n\n- `XAI_API_KEY` — X.AI API key for Grok's `x_search` tool. Declared in `requires:`, so it is **injected into this skill's environment** and is the **primary fetch path** for the narrative signal (step 1a). If it is ever unset, the skill degrades to WebSearch\u002FWebFetch at lower quality.\n- Notification channels configured via repo secrets (see CLAUDE.md).\n",{"name":5,"metadata":44,"description":7},{"mode":26,"tags":45,"title":46,"commits":47,"category":33,"requires":48,"permissions":50},[33,30],"Narrative Tracker",true,[49],"XAI_API_KEY",[51],"contents:write","main","504a21d53c0c9b485f6e90c43db83000825e1d3c","skills\u002Fnarrative-tracker",308,"https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon","https:\u002F\u002Fwww.aeon.fun\u002F","https:\u002F\u002Fgithub.com\u002Faeonfun","The most autonomous agent framework.","2026-11-15 03:40:18",{"source":8,"skillName":5,"commands":62,"prompt":72,"zipAvailable":47},[63,66,69],{"tool":64,"command":65},"npx","npx skills add aeonfun\u002Faeon --skill narrative-tracker",{"tool":67,"command":68},"bunx","bunx skills add aeonfun\u002Faeon --skill narrative-tracker",{"tool":70,"command":71},"pnpm","pnpm dlx skills add aeonfun\u002Faeon --skill narrative-tracker","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fskills\u002Fnarrative-tracker\u002FSKILL.md",[],[75,82,88,94,100,106],{"skillNo":76,"name":77,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":81},"01M7DSM3QWBJ44FKRMDDRAGNET","Warn Mixed Script","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fwarn-mixed-script\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":83,"name":84,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":81},"01M7DSM0496DRYYW5A4NAAHZDX","Warn Hidden CSS","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fwarn-hidden-css\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":81},"01M7DSKWCGDVYZSDJM31YHS2JB","Malicious Injection Uppercase","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fmalicious-injection-uppercase\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":81},"01M7DSKRV2RRA1V0HK13KYWJXD","Malicious Injection Homoglyph","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fmalicious-injection-homoglyph\u002FSKILL.md",[],[],{"skillNo":101,"name":102,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":81},"01M7DSKMXJKY0167YHVCKF7B49","Malicious Injection Disregard Prior","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fmalicious-injection-disregard-prior\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":81},"01M7DSKGFQ5V2VNSA4K97CVFMB","Benign Greek Math","https:\u002F\u002Fgithub.com\u002Faeonfun\u002Faeon\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fskill-scan-fixtures\u002Fbenign-greek-math\u002FSKILL.md",[],[]]