[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M718F46GB63WR5C3HWSWV0Z7":3,"skill-same:01M718F46GB63WR5C3HWSWV0Z7":85},{"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":31,"publishedAt":52,"linkOnly":53,"markdownContent":54,"markdownBody":55,"frontMatter":56,"refName":57,"commitSha":58,"skillDirectory":59,"forksCount":60,"repositoryHtmlUrl":61,"repositoryHomepageUrl":6,"ownerHtmlUrl":62,"ownerBlogUrl":63,"ownerDescription":6,"updatedAt":64,"install":65,"resources":78},"01M718F46GB63WR5C3HWSWV0Z7","tres-report-create",null,"Create (generate) any TRES Finance report end-to-end via the tres-mcp MCP server and GraphQL — trigger the export, verify a report row was actually created (guarding against silent failures), poll until done, and return the download link. Trigger this skill on ANY report creation request, including phrases like \"create a report\", \"generate a report\", \"export\", \"run a report\", \"I need a Transaction Ledger \u002F Balances \u002F Reconciliation \u002F Cost Basis \u002F Roll Forward \u002F Staking \u002F Audit report\", \"pull the data for\", \"download a report for\", \"give me a CSV\u002FXLSX of\", or whenever the user wants TRES to produce a report file. Use together with tres-report-advisor when the user is unsure which report they need.","anthropics\u002Fclaude-plugins-community","anthropics","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F76263028?v=4","01M697XGDPXJAVQ8391V2BCYTE","Anthropic","01M718C1RS4JJFM9KNJMTCAYJG",4892,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-report-create\u002FSKILL.md",[18,23,26],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000004","电子表格","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":27,"categoryName":28,"taxonomyType":29,"taxonomyTypeDesc":30},"01JBSEEDCAT000000000000034","行政与办公人员","PROFESSION","适用职业",[32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51],"tres report create","report","create","request","run","audit","advisor","generate","me","server","mcp server","done","when","MCP","end","xlsx","creation","balances","which","reconciliation","2026-11-10 10:12:30",false,"---\nname: tres-report-create\ndescription: Create (generate) any TRES Finance report end-to-end via the tres-mcp MCP server and GraphQL — trigger the export, verify a report row was actually created (guarding against silent failures), poll until done, and return the download link. Trigger this skill on ANY report creation request, including phrases like \"create a report\", \"generate a report\", \"export\", \"run a report\", \"I need a Transaction Ledger \u002F Balances \u002F Reconciliation \u002F Cost Basis \u002F Roll Forward \u002F Staking \u002F Audit report\", \"pull the data for\", \"download a report for\", \"give me a CSV\u002FXLSX of\", or whenever the user wants TRES to produce a report file. Use together with tres-report-advisor when the user is unsure which report they need.\n---\n\n# TRES Report Create\n\nYou create TRES Finance reports programmatically through the **tres-mcp** MCP server (GraphQL against the BFF). The flow is always: pick the report → trigger the export → poll until `DONE` → return the presigned `link`. Optionally download and analyze the file afterwards.\n\nIf you are unsure *which* report the user needs, use the `tres-report-advisor` skill first, then come back here to generate it.\n\n## Workflow\n\n```\n- [ ] Step 0: Clarify report type + date range + currency\n- [ ] Step 1: Trigger the export (GraphQL via MCP `execute`)\n- [ ] Step 1b: If the report isn't in the tables below, derive it from the schema\n- [ ] Step 1c: VERIFY a report row was actually created (poll by exact name) — catches silent failures\n- [ ] Step 2: Poll the report query until status == DONE\n- [ ] Step 3 (optional): Download + analyze the file\n```\n\n### Step 0 — Clarify scope (mandatory before triggering)\n\nConfirm before doing anything:\n\n- **Report type (required).** Match the user's wording to the tables below. If unsure, discover what's available:\n  ```graphql\n  query { availableReportTypes { name exportType entitiesType llmDescription } }\n  ```\n- **Date range (if applicable).** LEDGER \u002F AUDIT_LOG reports take `timestamp_Gte` \u002F `timestamp_Lte` in ISO 8601:\n  - \"Q1 2025\" → Gte `2025-01-01T00:00:00Z`, Lte `2025-03-31T23:59:59Z`\n  - \"2025\" → Gte `2025-01-01T00:00:00Z`, Lte `2025-12-31T23:59:59Z`\n  - If a date range is required and not given, **ask**.\n- **Currency.** Default `\"usd\"` unless the user says otherwise.\n- **Analysis goal (only if they want the data analyzed, not just the file).** What should the data answer? Default to a general summary (totals, top items, trends). Drives Step 3.\n\nConfirm briefly, e.g. *\"I'll generate a Transaction Ledger for Q1 2025 as CSV. Starting now.\"*\n\n### Step 1 — Trigger the export\n\nCall the GraphQL query that matches the report's `entitiesType`, passing the export parameters. Run it with the MCP `execute` tool (validate first with `validate_query` if unsure).\n\n**`entitiesType` → query to call**\n\n| entitiesType | Query |\n|---|---|\n| LEDGER | `transaction` |\n| ASSETS | `organizationBalance` |\n| BALANCE | `organizationBalance` |\n| HISTORICAL_BALANCE | `organizationBalance` |\n| ACCOUNTS | `internalAccount` |\n| GENERAL | `internalAccount` |\n| STAKING_DATA | `stakingYieldRecord` |\n| AUDIT_LOG | `auditLog` |\n| LOGIN_HISTORY | `loginHistoryExport` |\n\n**Report catalog (`exportFormat` values — verified against prod)**\n\n| Report name | exportFormat | entitiesType |\n|---|---|---|\n| Transaction Ledger | BASIC_RAW_TRANSACTIONS | LEDGER |\n| Realized Gains & Losses | EXTENDED_RAW_TRANSACTIONS | LEDGER |\n| Cost Breakdown Raw Transactions | COST_BREAKDOWN_RAW_TRANSACTIONS | LEDGER |\n| Rollup Breakdown † | ROLLUP_BREAKDOWN | LEDGER |\n| Ledger Reconciliation | RECONCILIATION | LEDGER |\n| Cost Basis Roll Forward | COST_BASIS_ROLL_FORWARD | LEDGER |\n| Asset Roll Forward | ASSET_ROLL_FORWARD | LEDGER |\n| ERP Pre-Sync | PRE_SYNC_JOURNAL | LEDGER |\n| ERP Post-Sync | POST_SYNC_JOURNAL | LEDGER |\n| MT940 Statement | MT940 | LEDGER |\n| Asset Balances | RAW_BALANCES | ASSETS |\n| Asset Balances V2 | RAW_BALANCES_V2 | ASSETS |\n| Asset Balances - Archives † | ARCHIVED_BALANCES | ASSETS |\n| Balance Trends | BALANCE_TRENDS | ASSETS |\n| Wallet Balances | INTERNAL_ACCOUNTS_BALANCES | ASSETS |\n| Cost Basis Stack Per Asset | COST_BASIS_STACK_PER_ASSET | ASSETS |\n| Asset Fiat Values | DAILY_ASSET_PRICING | ASSETS |\n| Revaluation Report | REEVALUATION | ASSETS |\n| Historical Balance Format | HISTORICAL_BALANCE | HISTORICAL_BALANCE |\n| Cost Basis Inventory † | COST_BASIS_INVENTORY | BALANCE |\n| Organization Wallets | INTERNAL_ACCOUNTS | ACCOUNTS |\n| Contacts | CONTACTS | ACCOUNTS |\n| Third Party Addresses | THIRD_PARTY_ADDRESSES | ACCOUNTS |\n| Chart of Account | CHART_OF_ACCOUNT | GENERAL |\n| ERP Rules | ERP_RULES | GENERAL |\n| Connected Custodians | CONNECTED_CUSTODIANS | GENERAL |\n| Staking Rewards & Positions | STAKING_DATA | STAKING_DATA |\n| Audit Trail \u002F Log | AUDIT_LOG | AUDIT_LOG |\n| Login History | LOGIN_HISTORY | LOGIN_HISTORY |\n\n† = special handling (see \"Reports needing extra parameters\" below). The org catalog evolves — `availableReportTypes` is authoritative. **If a requested report isn't in this table, go to Step 1b and derive it from the schema** rather than guessing.\n\n**Export parameters (always include):**\n\n- `exportFormat` — value from the table above (UPPER_CASE)\n- `exportName` — a descriptive, unique name (e.g. `\"Transaction Ledger Q1 2025\"`) — you'll match on this in Step 2\n- `currency` — `\"usd\"` unless told otherwise\n- `outputFormat` — `\"CSV\"` for downstream analysis, or `\"XLSX\"` if the user wants a spreadsheet\n\nAdd `timestamp_Gte` \u002F `timestamp_Lte` for date-ranged reports (LEDGER, AUDIT_LOG).\n\n> ### ⚠️ GraphQL variable types — get these EXACTLY right (first-run correctness)\n>\n> The BFF **strictly validates variable types**. A wrong type returns HTTP 400 and **no report is created** — and the `execute` tool surfaces this under an `error` \u002F `error_type` field, *not* the GraphQL `errors` array, so it is easy to miss. Declare variables with these exact types:\n>\n> | Variable | GraphQL type | Notes |\n> |---|---|---|\n> | `exportFormat` | `String` | the UPPER_CASE value |\n> | `exportName` | `String` | unique name |\n> | `currency` | `String` | e.g. `\"usd\"` |\n> | `outputFormat` | **`ReportOutputFormat`** | NOT `String`. Value `\"CSV\"` \u002F `\"XLSX\"` |\n> | `timestamp_Gte` \u002F `timestamp_Lte` | **`DateTime`** | NOT `String`. ISO 8601 value |\n> | `identifier_In`, `children_Asset_AssetClass_In`, `children_BelongsTo_In` | **`[String]`** | NOT `[ID]` |\n>\n> The params must be passed as **variables named exactly** `exportName`, `exportFormat`, etc. — the BFF reads them from `info.variable_values` by name. Inline literals or renamed variables (`$ef`) silently create **no** report.\n\nExample — trigger a Transaction Ledger export (verified working form):\n\n```graphql\nquery($limit: Int, $offset: Int, $timestamp_Gte: DateTime, $timestamp_Lte: DateTime,\n      $exportFormat: String, $exportName: String, $currency: String, $outputFormat: ReportOutputFormat) {\n  transaction(limit: $limit, offset: $offset, timestamp_Gte: $timestamp_Gte,\n              timestamp_Lte: $timestamp_Lte, exportFormat: $exportFormat,\n              exportName: $exportName, currency: $currency, outputFormat: $outputFormat) {\n    results { id }\n    totalCount\n  }\n}\n```\n\n```json\n{\n  \"limit\": 1,\n  \"offset\": 0,\n  \"timestamp_Gte\": \"2025-01-01T00:00:00Z\",\n  \"timestamp_Lte\": \"2025-03-31T23:59:59Z\",\n  \"exportFormat\": \"BASIC_RAW_TRANSACTIONS\",\n  \"exportName\": \"Transaction Ledger Q1 2025\",\n  \"currency\": \"usd\",\n  \"outputFormat\": \"CSV\"\n}\n```\n\nThe export then generates asynchronously in the background. A 200 with `totalCount` only confirms the *query* ran — it does **not** mean a report was created (verify in Step 1c) and it does **not** mean the file is ready (poll in Step 2).\n\n#### Reports needing extra parameters (omit these and the report errors or comes back empty)\n\n| Report (`exportFormat`) | Trigger via | Required extra params |\n|---|---|---|\n| `ROLLUP_BREAKDOWN` | `transaction` | `identifier_In: [String]` — rollup parent hashes to decompose. Without it the report is **empty**. |\n| `COST_BASIS_INVENTORY` | `transaction` (not `organizationBalance`) | exactly **one** `children_Asset_AssetClass_In: [String]` (asset-class id). Optional: at most one `children_BelongsTo_In: [String]`. Missing\u002Fmultiple → report finishes with `status: ERROR` (\"must select exactly one asset class\"). Get an asset-class id from `query { assetClass(limit: N) { results { id name symbol } } }`. |\n| `ARCHIVED_BALANCES` | `organizationBalance` | Requires an existing **commit snapshot** for the target date. If none exists it stays pending — prefer `HISTORICAL_BALANCE` (Time Capsule) for arbitrary dates. |\n\n**Point-in-time balances.** For a balance report *as of a past date*, `organizationBalance` also accepts `balanceDate: DateTime` (reconstructs balances at that date — this is the Time Capsule path; passing `RAW_BALANCES` + `balanceDate` is promoted to `HISTORICAL_BALANCE`) and `commitId: UUID` (pull from a specific commit snapshot). Omit both for current balances.\n\n#### `LOGIN_HISTORY` has a different shape\n\n`loginHistoryExport` takes only `exportFormat: String!`, `exportName: String!`, `outputFormat: ReportOutputFormat` (no currency\u002Flimit\u002Ftimestamps) and returns `{ success reportId }`:\n\n```graphql\nquery($exportFormat: String!, $exportName: String!, $outputFormat: ReportOutputFormat) {\n  loginHistoryExport(exportFormat: $exportFormat, exportName: $exportName, outputFormat: $outputFormat) {\n    success\n    reportId\n  }\n}\n```\n\n### Step 1b — Unknown or newly added report? Derive it from the schema (self-service)\n\n`availableReportTypes` is the source of truth — it lists **every** report the org supports, including ones not in the tables above. When a requested report has no row here, do not guess: discover the trigger from the live schema using the MCP introspection tools.\n\n```\n- [ ] 1. availableReportTypes -> get this report's exportType + entitiesType + llmDescription\n- [ ] 2. Pick the trigger query from entitiesType (Step 1 table). If entitiesType is new\u002Funclear,\n         default: transaction (tx \u002F ledger \u002F cost-basis), organizationBalance (balances \u002F assets),\n         internalAccount (accounts \u002F metadata).\n- [ ] 3. introspect(\u003Cquery>) -> read the EXACT arg names AND types\n- [ ] 4. build_query(\u003Cquery>) -> canonical variables-as-args template (optional, handy)\n- [ ] 5. validate_query(\u003Cassembled query>) BEFORE executing\n- [ ] 6. Trigger with the 4 export params (+ timestamps only if the query exposes them)\n- [ ] 7. Poll; if it errors, apply the error->fix table, then re-trigger\n- [ ] 8. On DONE: return the link AND record the working recipe\n```\n\nPrinciples that make this work first-try:\n\n1. **Read the format, never invent it.** Use `exportType` from `availableReportTypes` verbatim as `exportFormat`.\n2. **Derive variable types from `introspect`, not from memory.** Declare each variable with the exact type the schema reports. This alone prevents the 400 \"used in position expecting type X\" failures. Sanity-check the usual ones: `outputFormat` → `ReportOutputFormat`, `currency` → `String`, `timestamp_*` → `DateTime`, list filters → `[String]`.\n3. **Validate before executing.** `validate_query` catches arg\u002Ftype\u002Fselection mistakes without creating a junk report.\n4. **Start minimal, then add only what the error demands.** Trigger with just the export params first; let the error messages tell you which extra filter a generator requires.\n\n**Error → fix table** (covers every failure mode seen against prod):\n\n| Error | Meaning | Fix |\n|---|---|---|\n| `Variable '$X' of type 'A' used in position expecting type 'B'` | wrong declared type | re-declare `$X` as `B`, re-validate |\n| `Unknown argument 'X' on field 'Y'` | that filter isn't on this query | switch to a query that exposes it (often `transaction`) or drop it |\n| `Variable '$X' is never used` | declared a var with no matching arg | make it a real arg on the query, or remove it |\n| 200 trigger, but report `status: ERROR` (\"must select exactly one …\" \u002F \"required …\") | generator needs an extra required filter | `introspect` to find the matching arg name, fetch valid values from its list query (e.g. `assetClass`, `internalAccount`), add exactly what's required, re-trigger |\n| report stays `IN_PROGRESS` indefinitely | data prerequisite missing (e.g. no commit snapshot) | tell the user what's missing and suggest an alternative report |\n\n**Persist what you learn.** Once a new report triggers cleanly to `DONE`:\n- Short term: save the recipe via the MCP `memory` tool — trigger query, `exportFormat`, working variable types, and any required filters.\n- Durable: propose a new row for this skill's tables (report name, `exportFormat`, trigger query, required params) and surface it to the user so the skill stays current.\n\n### Step 1c — Verify a report row was created (MANDATORY anti-silent-failure check)\n\n**Do this immediately after every trigger, before you tell the user anything succeeded.** The trigger is a query *side effect*: if the request is even slightly malformed (inline literals instead of variables, a renamed variable like `$ef`, a missing required param), the BFF returns `200` with `errors: null` and **silently creates no report**. The only reliable way to know it worked is to look the row up by name.\n\nQuery for the row using the **exact** `exportName` you triggered with:\n\n```graphql\nquery($name: String, $ordering: String, $limit: Int) {\n  report(name: $name, ordering: $ordering, limit: $limit) {\n    results { id name status }\n  }\n}\n```\n\n```json\n{ \"name\": \"\u003Cthe exact exportName from Step 1>\", \"ordering\": \"-created_at\", \"limit\": 1 }\n```\n\n- **`results` is empty (no row)** → this is a **SILENT FAILURE**, not \"in progress\". Do **NOT** report success. Re-check the trigger against the rules above — params passed as **variables** (not inline literals), variables **named exactly** `exportName`\u002F`exportFormat`\u002F`currency`\u002F`outputFormat`\u002F`timestamp_*`, correct types — then re-trigger **once**. If it still creates no row, stop and tell the user it failed silently and could not be created (don't loop forever).\n- **A row exists** → good, the report was created. Continue to Step 2 to poll it to `DONE`.\n\n> Make the `exportName` unique per run (e.g. append the date or a timestamp) so this lookup matches *your* row and not an older report with the same name.\n\n### Step 2 — Poll until done\n\nFind the report by the exact `exportName` from Step 1 (most reliable match):\n\n```graphql\nquery($name: String, $ordering: String, $limit: Int) {\n  report(name: $name, ordering: $ordering, limit: $limit) {\n    results { id name status link reportSize exportFormat createdAt }\n  }\n}\n```\n\n```json\n{ \"name\": \"\u003Cthe exportName from Step 1>\", \"ordering\": \"-created_at\", \"limit\": 1 }\n```\n\n**Field\u002Farg gotchas (these fail silently — get them right):**\n\n- The ordering arg is `ordering`, **not** `orderBy`.\n- The download field is `link` (a presigned S3 URL), **not** `downloadUrl`.\n- The report-type field is `exportFormat`, **not** `exportType`.\n- Do **not** filter by `exportFormat` to find your report: the stored value is **lowercase** (e.g. `rollup_breakdown`) while you triggered with UPPER_CASE, so an exact filter matches nothing. Filter by `name`.\n\n**Polling loop:**\n\n1. Query for the latest report by name.\n2. Check `status`:\n   - `IN_PROGRESS` → wait ~60s (e.g. `sleep 60` in a shell), then query again.\n   - `DONE` → `link` holds the presigned download URL. Proceed.\n   - `ERROR` → the generator rejected the inputs (e.g. a missing required filter). Read the failure message, fix the params, and re-trigger. Do not silently retry the same call.\n3. Repeat up to **10 times**, waiting ~60s between attempts.\n\n> **Check BOTH error channels.** A report can fail in two ways: (1) the trigger returns HTTP 400 with the problem under `error` \u002F `error_type` (and an empty\u002Fabsent GraphQL `errors`) — treat any non-empty `error` as a hard failure, do **not** report success; (2) the trigger returns 200 but the report later finishes with `status: ERROR`. Never assume success from a 200 alone.\n\nSmall reports finish in seconds; large reports (many assets, wide date ranges, large orgs) can take minutes to hours — set expectations early. Once `status == DONE`, share the `link` with the user.\n\n**Presenting the download link.** Never paste the raw presigned URL into the chat — it's long, ugly, and full of credentials. Always render it as a clean clickable **markdown link** with friendly anchor text, putting the full URL in the link target:\n\n```markdown\n[Click here to download your report](\u003Cthe full presigned link>)\n```\n\nYou may tailor the anchor text to the report, e.g. `[Download your ERP Pre-Sync report (CSV)](\u003Clink>)`. Keep the anchor short and human; the URL itself stays hidden behind it. Mention that the link is presigned and expires after a limited window (~24h).\n\n### Step 3 — Download + analyze (optional)\n\nIf the user wants the data analyzed (not just the link), download via the presigned URL and inspect it locally with pandas:\n\n```python\nimport pandas as pd, requests\nfrom io import BytesIO\n\nresp = requests.get(\"\u003Cpresigned_url>\")\nresp.raise_for_status()\ndf = pd.read_csv(BytesIO(resp.content))\nprint(df.shape, list(df.columns))\nprint(df.head())\nprint(df.dtypes)\n```\n\nAlways start with `df.head()` and `df.dtypes`, then run the requested analysis (`groupby`, `value_counts`, `describe`, `pivot_table`). Format money as `f\"${value:,.2f}\"` and summarize large frames rather than dumping every row. For deeper per-report analysis guidance, use the `tres-report-analyzer` skill.\n\n## Rules\n\n- **ALWAYS** confirm report type (and date range, if required) before triggering.\n- Match the query to the report's `entitiesType` using the table above — **except** the special-case reports above (`COST_BASIS_INVENTORY` triggers via `transaction`).\n- For any report NOT in the tables, follow **Step 1b**: `introspect` the query for exact arg types, `validate_query`, trigger minimal, fix from the error→fix table, then persist the recipe. Never hand-guess types or filters for an unknown report.\n- **Variable types are strict:** `outputFormat` is `ReportOutputFormat`, `timestamp_*` are `DateTime`, list filters are `[String]`. A wrong type = HTTP 400 = no report. Pass params as variables named exactly `exportName` \u002F `exportFormat` \u002F etc.\n- **ALWAYS run Step 1c after triggering.** Verify the report row exists by exact `exportName` before claiming anything worked. An empty lookup means a **silent failure** (200 + `errors: null` + no row), not \"in progress\" — never report success without a confirmed row. Use a unique `exportName` per run so the lookup matches your row.\n- For `ROLLUP_BREAKDOWN`, you MUST pass `identifier_In`; for `COST_BASIS_INVENTORY`, exactly one `children_Asset_AssetClass_In` — otherwise the report is empty or ERRORs.\n- After triggering, verify success on **both** channels: no `error`\u002F`error_type` on the trigger, and `status` reaches `DONE` (not `ERROR`).\n- Poll the `report` query by `name`, not `exportFormat`; use `ordering` (not `orderBy`) and `link` (not `downloadUrl`).\n- **Present the download link as a clickable markdown link** (e.g. `[Click here to download your report](\u003Clink>)`), never the raw presigned URL. Note that the link is presigned and expires after a limited window.\n- Default `currency` to `\"usd\"` and `outputFormat` to `\"CSV\"` unless the user specifies otherwise.\n- Never truncate or shorten transaction hashes, addresses, report IDs, or `identifier_In` values — always show\u002Fpass them in full.\n- This is a read\u002Fexport skill — **NEVER** run mutations that modify platform data.\n","# TRES Report Create\n\nYou create TRES Finance reports programmatically through the **tres-mcp** MCP server (GraphQL against the BFF). The flow is always: pick the report → trigger the export → poll until `DONE` → return the presigned `link`. Optionally download and analyze the file afterwards.\n\nIf you are unsure *which* report the user needs, use the `tres-report-advisor` skill first, then come back here to generate it.\n\n## Workflow\n\n```\n- [ ] Step 0: Clarify report type + date range + currency\n- [ ] Step 1: Trigger the export (GraphQL via MCP `execute`)\n- [ ] Step 1b: If the report isn't in the tables below, derive it from the schema\n- [ ] Step 1c: VERIFY a report row was actually created (poll by exact name) — catches silent failures\n- [ ] Step 2: Poll the report query until status == DONE\n- [ ] Step 3 (optional): Download + analyze the file\n```\n\n### Step 0 — Clarify scope (mandatory before triggering)\n\nConfirm before doing anything:\n\n- **Report type (required).** Match the user's wording to the tables below. If unsure, discover what's available:\n  ```graphql\n  query { availableReportTypes { name exportType entitiesType llmDescription } }\n  ```\n- **Date range (if applicable).** LEDGER \u002F AUDIT_LOG reports take `timestamp_Gte` \u002F `timestamp_Lte` in ISO 8601:\n  - \"Q1 2025\" → Gte `2025-01-01T00:00:00Z`, Lte `2025-03-31T23:59:59Z`\n  - \"2025\" → Gte `2025-01-01T00:00:00Z`, Lte `2025-12-31T23:59:59Z`\n  - If a date range is required and not given, **ask**.\n- **Currency.** Default `\"usd\"` unless the user says otherwise.\n- **Analysis goal (only if they want the data analyzed, not just the file).** What should the data answer? Default to a general summary (totals, top items, trends). Drives Step 3.\n\nConfirm briefly, e.g. *\"I'll generate a Transaction Ledger for Q1 2025 as CSV. Starting now.\"*\n\n### Step 1 — Trigger the export\n\nCall the GraphQL query that matches the report's `entitiesType`, passing the export parameters. Run it with the MCP `execute` tool (validate first with `validate_query` if unsure).\n\n**`entitiesType` → query to call**\n\n| entitiesType | Query |\n|---|---|\n| LEDGER | `transaction` |\n| ASSETS | `organizationBalance` |\n| BALANCE | `organizationBalance` |\n| HISTORICAL_BALANCE | `organizationBalance` |\n| ACCOUNTS | `internalAccount` |\n| GENERAL | `internalAccount` |\n| STAKING_DATA | `stakingYieldRecord` |\n| AUDIT_LOG | `auditLog` |\n| LOGIN_HISTORY | `loginHistoryExport` |\n\n**Report catalog (`exportFormat` values — verified against prod)**\n\n| Report name | exportFormat | entitiesType |\n|---|---|---|\n| Transaction Ledger | BASIC_RAW_TRANSACTIONS | LEDGER |\n| Realized Gains & Losses | EXTENDED_RAW_TRANSACTIONS | LEDGER |\n| Cost Breakdown Raw Transactions | COST_BREAKDOWN_RAW_TRANSACTIONS | LEDGER |\n| Rollup Breakdown † | ROLLUP_BREAKDOWN | LEDGER |\n| Ledger Reconciliation | RECONCILIATION | LEDGER |\n| Cost Basis Roll Forward | COST_BASIS_ROLL_FORWARD | LEDGER |\n| Asset Roll Forward | ASSET_ROLL_FORWARD | LEDGER |\n| ERP Pre-Sync | PRE_SYNC_JOURNAL | LEDGER |\n| ERP Post-Sync | POST_SYNC_JOURNAL | LEDGER |\n| MT940 Statement | MT940 | LEDGER |\n| Asset Balances | RAW_BALANCES | ASSETS |\n| Asset Balances V2 | RAW_BALANCES_V2 | ASSETS |\n| Asset Balances - Archives † | ARCHIVED_BALANCES | ASSETS |\n| Balance Trends | BALANCE_TRENDS | ASSETS |\n| Wallet Balances | INTERNAL_ACCOUNTS_BALANCES | ASSETS |\n| Cost Basis Stack Per Asset | COST_BASIS_STACK_PER_ASSET | ASSETS |\n| Asset Fiat Values | DAILY_ASSET_PRICING | ASSETS |\n| Revaluation Report | REEVALUATION | ASSETS |\n| Historical Balance Format | HISTORICAL_BALANCE | HISTORICAL_BALANCE |\n| Cost Basis Inventory † | COST_BASIS_INVENTORY | BALANCE |\n| Organization Wallets | INTERNAL_ACCOUNTS | ACCOUNTS |\n| Contacts | CONTACTS | ACCOUNTS |\n| Third Party Addresses | THIRD_PARTY_ADDRESSES | ACCOUNTS |\n| Chart of Account | CHART_OF_ACCOUNT | GENERAL |\n| ERP Rules | ERP_RULES | GENERAL |\n| Connected Custodians | CONNECTED_CUSTODIANS | GENERAL |\n| Staking Rewards & Positions | STAKING_DATA | STAKING_DATA |\n| Audit Trail \u002F Log | AUDIT_LOG | AUDIT_LOG |\n| Login History | LOGIN_HISTORY | LOGIN_HISTORY |\n\n† = special handling (see \"Reports needing extra parameters\" below). The org catalog evolves — `availableReportTypes` is authoritative. **If a requested report isn't in this table, go to Step 1b and derive it from the schema** rather than guessing.\n\n**Export parameters (always include):**\n\n- `exportFormat` — value from the table above (UPPER_CASE)\n- `exportName` — a descriptive, unique name (e.g. `\"Transaction Ledger Q1 2025\"`) — you'll match on this in Step 2\n- `currency` — `\"usd\"` unless told otherwise\n- `outputFormat` — `\"CSV\"` for downstream analysis, or `\"XLSX\"` if the user wants a spreadsheet\n\nAdd `timestamp_Gte` \u002F `timestamp_Lte` for date-ranged reports (LEDGER, AUDIT_LOG).\n\n> ### ⚠️ GraphQL variable types — get these EXACTLY right (first-run correctness)\n>\n> The BFF **strictly validates variable types**. A wrong type returns HTTP 400 and **no report is created** — and the `execute` tool surfaces this under an `error` \u002F `error_type` field, *not* the GraphQL `errors` array, so it is easy to miss. Declare variables with these exact types:\n>\n> | Variable | GraphQL type | Notes |\n> |---|---|---|\n> | `exportFormat` | `String` | the UPPER_CASE value |\n> | `exportName` | `String` | unique name |\n> | `currency` | `String` | e.g. `\"usd\"` |\n> | `outputFormat` | **`ReportOutputFormat`** | NOT `String`. Value `\"CSV\"` \u002F `\"XLSX\"` |\n> | `timestamp_Gte` \u002F `timestamp_Lte` | **`DateTime`** | NOT `String`. ISO 8601 value |\n> | `identifier_In`, `children_Asset_AssetClass_In`, `children_BelongsTo_In` | **`[String]`** | NOT `[ID]` |\n>\n> The params must be passed as **variables named exactly** `exportName`, `exportFormat`, etc. — the BFF reads them from `info.variable_values` by name. Inline literals or renamed variables (`$ef`) silently create **no** report.\n\nExample — trigger a Transaction Ledger export (verified working form):\n\n```graphql\nquery($limit: Int, $offset: Int, $timestamp_Gte: DateTime, $timestamp_Lte: DateTime,\n      $exportFormat: String, $exportName: String, $currency: String, $outputFormat: ReportOutputFormat) {\n  transaction(limit: $limit, offset: $offset, timestamp_Gte: $timestamp_Gte,\n              timestamp_Lte: $timestamp_Lte, exportFormat: $exportFormat,\n              exportName: $exportName, currency: $currency, outputFormat: $outputFormat) {\n    results { id }\n    totalCount\n  }\n}\n```\n\n```json\n{\n  \"limit\": 1,\n  \"offset\": 0,\n  \"timestamp_Gte\": \"2025-01-01T00:00:00Z\",\n  \"timestamp_Lte\": \"2025-03-31T23:59:59Z\",\n  \"exportFormat\": \"BASIC_RAW_TRANSACTIONS\",\n  \"exportName\": \"Transaction Ledger Q1 2025\",\n  \"currency\": \"usd\",\n  \"outputFormat\": \"CSV\"\n}\n```\n\nThe export then generates asynchronously in the background. A 200 with `totalCount` only confirms the *query* ran — it does **not** mean a report was created (verify in Step 1c) and it does **not** mean the file is ready (poll in Step 2).\n\n#### Reports needing extra parameters (omit these and the report errors or comes back empty)\n\n| Report (`exportFormat`) | Trigger via | Required extra params |\n|---|---|---|\n| `ROLLUP_BREAKDOWN` | `transaction` | `identifier_In: [String]` — rollup parent hashes to decompose. Without it the report is **empty**. |\n| `COST_BASIS_INVENTORY` | `transaction` (not `organizationBalance`) | exactly **one** `children_Asset_AssetClass_In: [String]` (asset-class id). Optional: at most one `children_BelongsTo_In: [String]`. Missing\u002Fmultiple → report finishes with `status: ERROR` (\"must select exactly one asset class\"). Get an asset-class id from `query { assetClass(limit: N) { results { id name symbol } } }`. |\n| `ARCHIVED_BALANCES` | `organizationBalance` | Requires an existing **commit snapshot** for the target date. If none exists it stays pending — prefer `HISTORICAL_BALANCE` (Time Capsule) for arbitrary dates. |\n\n**Point-in-time balances.** For a balance report *as of a past date*, `organizationBalance` also accepts `balanceDate: DateTime` (reconstructs balances at that date — this is the Time Capsule path; passing `RAW_BALANCES` + `balanceDate` is promoted to `HISTORICAL_BALANCE`) and `commitId: UUID` (pull from a specific commit snapshot). Omit both for current balances.\n\n#### `LOGIN_HISTORY` has a different shape\n\n`loginHistoryExport` takes only `exportFormat: String!`, `exportName: String!`, `outputFormat: ReportOutputFormat` (no currency\u002Flimit\u002Ftimestamps) and returns `{ success reportId }`:\n\n```graphql\nquery($exportFormat: String!, $exportName: String!, $outputFormat: ReportOutputFormat) {\n  loginHistoryExport(exportFormat: $exportFormat, exportName: $exportName, outputFormat: $outputFormat) {\n    success\n    reportId\n  }\n}\n```\n\n### Step 1b — Unknown or newly added report? Derive it from the schema (self-service)\n\n`availableReportTypes` is the source of truth — it lists **every** report the org supports, including ones not in the tables above. When a requested report has no row here, do not guess: discover the trigger from the live schema using the MCP introspection tools.\n\n```\n- [ ] 1. availableReportTypes -> get this report's exportType + entitiesType + llmDescription\n- [ ] 2. Pick the trigger query from entitiesType (Step 1 table). If entitiesType is new\u002Funclear,\n         default: transaction (tx \u002F ledger \u002F cost-basis), organizationBalance (balances \u002F assets),\n         internalAccount (accounts \u002F metadata).\n- [ ] 3. introspect(\u003Cquery>) -> read the EXACT arg names AND types\n- [ ] 4. build_query(\u003Cquery>) -> canonical variables-as-args template (optional, handy)\n- [ ] 5. validate_query(\u003Cassembled query>) BEFORE executing\n- [ ] 6. Trigger with the 4 export params (+ timestamps only if the query exposes them)\n- [ ] 7. Poll; if it errors, apply the error->fix table, then re-trigger\n- [ ] 8. On DONE: return the link AND record the working recipe\n```\n\nPrinciples that make this work first-try:\n\n1. **Read the format, never invent it.** Use `exportType` from `availableReportTypes` verbatim as `exportFormat`.\n2. **Derive variable types from `introspect`, not from memory.** Declare each variable with the exact type the schema reports. This alone prevents the 400 \"used in position expecting type X\" failures. Sanity-check the usual ones: `outputFormat` → `ReportOutputFormat`, `currency` → `String`, `timestamp_*` → `DateTime`, list filters → `[String]`.\n3. **Validate before executing.** `validate_query` catches arg\u002Ftype\u002Fselection mistakes without creating a junk report.\n4. **Start minimal, then add only what the error demands.** Trigger with just the export params first; let the error messages tell you which extra filter a generator requires.\n\n**Error → fix table** (covers every failure mode seen against prod):\n\n| Error | Meaning | Fix |\n|---|---|---|\n| `Variable '$X' of type 'A' used in position expecting type 'B'` | wrong declared type | re-declare `$X` as `B`, re-validate |\n| `Unknown argument 'X' on field 'Y'` | that filter isn't on this query | switch to a query that exposes it (often `transaction`) or drop it |\n| `Variable '$X' is never used` | declared a var with no matching arg | make it a real arg on the query, or remove it |\n| 200 trigger, but report `status: ERROR` (\"must select exactly one …\" \u002F \"required …\") | generator needs an extra required filter | `introspect` to find the matching arg name, fetch valid values from its list query (e.g. `assetClass`, `internalAccount`), add exactly what's required, re-trigger |\n| report stays `IN_PROGRESS` indefinitely | data prerequisite missing (e.g. no commit snapshot) | tell the user what's missing and suggest an alternative report |\n\n**Persist what you learn.** Once a new report triggers cleanly to `DONE`:\n- Short term: save the recipe via the MCP `memory` tool — trigger query, `exportFormat`, working variable types, and any required filters.\n- Durable: propose a new row for this skill's tables (report name, `exportFormat`, trigger query, required params) and surface it to the user so the skill stays current.\n\n### Step 1c — Verify a report row was created (MANDATORY anti-silent-failure check)\n\n**Do this immediately after every trigger, before you tell the user anything succeeded.** The trigger is a query *side effect*: if the request is even slightly malformed (inline literals instead of variables, a renamed variable like `$ef`, a missing required param), the BFF returns `200` with `errors: null` and **silently creates no report**. The only reliable way to know it worked is to look the row up by name.\n\nQuery for the row using the **exact** `exportName` you triggered with:\n\n```graphql\nquery($name: String, $ordering: String, $limit: Int) {\n  report(name: $name, ordering: $ordering, limit: $limit) {\n    results { id name status }\n  }\n}\n```\n\n```json\n{ \"name\": \"\u003Cthe exact exportName from Step 1>\", \"ordering\": \"-created_at\", \"limit\": 1 }\n```\n\n- **`results` is empty (no row)** → this is a **SILENT FAILURE**, not \"in progress\". Do **NOT** report success. Re-check the trigger against the rules above — params passed as **variables** (not inline literals), variables **named exactly** `exportName`\u002F`exportFormat`\u002F`currency`\u002F`outputFormat`\u002F`timestamp_*`, correct types — then re-trigger **once**. If it still creates no row, stop and tell the user it failed silently and could not be created (don't loop forever).\n- **A row exists** → good, the report was created. Continue to Step 2 to poll it to `DONE`.\n\n> Make the `exportName` unique per run (e.g. append the date or a timestamp) so this lookup matches *your* row and not an older report with the same name.\n\n### Step 2 — Poll until done\n\nFind the report by the exact `exportName` from Step 1 (most reliable match):\n\n```graphql\nquery($name: String, $ordering: String, $limit: Int) {\n  report(name: $name, ordering: $ordering, limit: $limit) {\n    results { id name status link reportSize exportFormat createdAt }\n  }\n}\n```\n\n```json\n{ \"name\": \"\u003Cthe exportName from Step 1>\", \"ordering\": \"-created_at\", \"limit\": 1 }\n```\n\n**Field\u002Farg gotchas (these fail silently — get them right):**\n\n- The ordering arg is `ordering`, **not** `orderBy`.\n- The download field is `link` (a presigned S3 URL), **not** `downloadUrl`.\n- The report-type field is `exportFormat`, **not** `exportType`.\n- Do **not** filter by `exportFormat` to find your report: the stored value is **lowercase** (e.g. `rollup_breakdown`) while you triggered with UPPER_CASE, so an exact filter matches nothing. Filter by `name`.\n\n**Polling loop:**\n\n1. Query for the latest report by name.\n2. Check `status`:\n   - `IN_PROGRESS` → wait ~60s (e.g. `sleep 60` in a shell), then query again.\n   - `DONE` → `link` holds the presigned download URL. Proceed.\n   - `ERROR` → the generator rejected the inputs (e.g. a missing required filter). Read the failure message, fix the params, and re-trigger. Do not silently retry the same call.\n3. Repeat up to **10 times**, waiting ~60s between attempts.\n\n> **Check BOTH error channels.** A report can fail in two ways: (1) the trigger returns HTTP 400 with the problem under `error` \u002F `error_type` (and an empty\u002Fabsent GraphQL `errors`) — treat any non-empty `error` as a hard failure, do **not** report success; (2) the trigger returns 200 but the report later finishes with `status: ERROR`. Never assume success from a 200 alone.\n\nSmall reports finish in seconds; large reports (many assets, wide date ranges, large orgs) can take minutes to hours — set expectations early. Once `status == DONE`, share the `link` with the user.\n\n**Presenting the download link.** Never paste the raw presigned URL into the chat — it's long, ugly, and full of credentials. Always render it as a clean clickable **markdown link** with friendly anchor text, putting the full URL in the link target:\n\n```markdown\n[Click here to download your report](\u003Cthe full presigned link>)\n```\n\nYou may tailor the anchor text to the report, e.g. `[Download your ERP Pre-Sync report (CSV)](\u003Clink>)`. Keep the anchor short and human; the URL itself stays hidden behind it. Mention that the link is presigned and expires after a limited window (~24h).\n\n### Step 3 — Download + analyze (optional)\n\nIf the user wants the data analyzed (not just the link), download via the presigned URL and inspect it locally with pandas:\n\n```python\nimport pandas as pd, requests\nfrom io import BytesIO\n\nresp = requests.get(\"\u003Cpresigned_url>\")\nresp.raise_for_status()\ndf = pd.read_csv(BytesIO(resp.content))\nprint(df.shape, list(df.columns))\nprint(df.head())\nprint(df.dtypes)\n```\n\nAlways start with `df.head()` and `df.dtypes`, then run the requested analysis (`groupby`, `value_counts`, `describe`, `pivot_table`). Format money as `f\"${value:,.2f}\"` and summarize large frames rather than dumping every row. For deeper per-report analysis guidance, use the `tres-report-analyzer` skill.\n\n## Rules\n\n- **ALWAYS** confirm report type (and date range, if required) before triggering.\n- Match the query to the report's `entitiesType` using the table above — **except** the special-case reports above (`COST_BASIS_INVENTORY` triggers via `transaction`).\n- For any report NOT in the tables, follow **Step 1b**: `introspect` the query for exact arg types, `validate_query`, trigger minimal, fix from the error→fix table, then persist the recipe. Never hand-guess types or filters for an unknown report.\n- **Variable types are strict:** `outputFormat` is `ReportOutputFormat`, `timestamp_*` are `DateTime`, list filters are `[String]`. A wrong type = HTTP 400 = no report. Pass params as variables named exactly `exportName` \u002F `exportFormat` \u002F etc.\n- **ALWAYS run Step 1c after triggering.** Verify the report row exists by exact `exportName` before claiming anything worked. An empty lookup means a **silent failure** (200 + `errors: null` + no row), not \"in progress\" — never report success without a confirmed row. Use a unique `exportName` per run so the lookup matches your row.\n- For `ROLLUP_BREAKDOWN`, you MUST pass `identifier_In`; for `COST_BASIS_INVENTORY`, exactly one `children_Asset_AssetClass_In` — otherwise the report is empty or ERRORs.\n- After triggering, verify success on **both** channels: no `error`\u002F`error_type` on the trigger, and `status` reaches `DONE` (not `ERROR`).\n- Poll the `report` query by `name`, not `exportFormat`; use `ordering` (not `orderBy`) and `link` (not `downloadUrl`).\n- **Present the download link as a clickable markdown link** (e.g. `[Click here to download your report](\u003Clink>)`), never the raw presigned URL. Note that the link is presigned and expires after a limited window.\n- Default `currency` to `\"usd\"` and `outputFormat` to `\"CSV\"` unless the user specifies otherwise.\n- Never truncate or shorten transaction hashes, addresses, report IDs, or `identifier_In` values — always show\u002Fpass them in full.\n- This is a read\u002Fexport skill — **NEVER** run mutations that modify platform data.\n",{"name":5,"description":7},"main","163301d1904c0a994539f5226ce0fcd65b8652bf","tres-finance-plugin\u002Fskills\u002Ftres-report-create",363,"https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community","https:\u002F\u002Fgithub.com\u002Fanthropics","https:\u002F\u002Fanthropic.com","2026-11-12 15:56:31",{"source":8,"skillName":5,"commands":66,"prompt":76,"zipAvailable":77},[67,70,73],{"tool":68,"command":69},"npx","npx skills add anthropics\u002Fclaude-plugins-community --skill tres-report-create",{"tool":71,"command":72},"bunx","bunx skills add anthropics\u002Fclaude-plugins-community --skill tres-report-create",{"tool":74,"command":75},"pnpm","pnpm dlx skills add anthropics\u002Fclaude-plugins-community --skill tres-report-create","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-report-create\u002FSKILL.md",true,[79],{"path":80,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":83,"sourceHtmlUrl":84},"tests\u002Frun_report_matrix.py","SCRIPT","脚本",17222,"https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-report-create\u002Ftests\u002Frun_report_matrix.py",[86,114,139,161,185,208],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":97,"publishedAt":52},"01M718G9QZ4B2DJVEB3MAW5ZB1","tres-wallets-upload","Upload and onboard multiple on-chain wallets or exchange accounts into Tres Finance. Use this skill whenever the user wants to add wallets, onboard wallets, import wallets, upload wallet addresses, ingest wallets, or connect exchanges into Tres — whether from a file (CSV\u002FExcel) or by typing them manually. This skill guides Claude through the full flow: wallet type selection → input collection → validation → preview with editable HTML table → exchange credential collection (if exchange) → user confirmation → batch creation via the TRES MCP API. Always trigger this skill for wallet onboarding, wallet data ingestion, or any request to \"add wallets to Tres\".","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-wallets-upload\u002FSKILL.md",[92,93,94],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":95,"categoryName":96,"taxonomyType":29,"taxonomyTypeDesc":30},"01JBSEEDCAT000000000000040","财务",[98,99,100,101,102,103,104,48,47,105,45,106,107,108,109,110,111,112,113,35],"wallets","tres wallets upload","upload","typing","ingestion","Wallet","chain","selection","confirmation","connect","collection","preview","api","batch","guides","on-chain",{"skillNo":115,"name":116,"nameZh":6,"description":117,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":118,"categories":119,"keywords":120,"publishedAt":52},"01M718G5RD9064Z574RQ4SEY38","tres-upload-tx-header-validation","The exact column header names required by the TRES Finance bulk transaction CSV upload. Use whenever building, fixing, or validating the header row of a TRES manual transaction or sub-transaction upload CSV — especially after an upload fails with \"Missing required column(s): ...\". Trigger phrases include: 'fix my CSV headers', 'bulk transaction upload failed', 'missing required columns', 'validate TRES upload CSV', 'transaction CSV template headers'. This skill is ONLY about the header naming convention; it does not cover row values or upload workflow. Do NOT trigger for contacts import CSV (use tres-import-contacts), wallet upload CSV (use tres-wallets-upload), or general CSV parsing unrelated to bulk TX upload headers.","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-upload-tx-header-validation\u002FSKILL.md",[],[121,100,122,123,124,125,103,126,127,128,129,130,131,132,133,134,135,136,137,138],"tres upload tx header validation","validation","bulk","do","import","parsing","for","manual","trigger","fix","missing","general","only","use","workflow","csv","contacts","with",{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":146,"publishedAt":52},"01M718G1F3DBWTX5H9RWPHHB4F","tres-tx-story","Analyze a blockchain transaction by its hash using TRES Finance MCP and generate a visual flow diagram showing all asset movements, followed by a plain-language explanation of what happened. Use this skill whenever the user provides a transaction hash and asks to \"analyze\", \"explain\", \"visualize\", \"diagram\", \"show me what happened in\", or \"tell the story of\" a transaction. Also trigger when the user pastes a tx hash (long hex string starting with 0x) and asks anything about it. Always use this skill for transaction hash analysis — do not attempt to answer from memory alone.","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-tx-story\u002FSKILL.md",[145],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[147,148,149,150,151,45,44,152,153,154,40,155,39,156,157,158,159,124,160,127],"tres tx story","Story","show","memory","hex","show me","analysis","analyze","diagram","flow","asset","blockchain","STRING","visualize",{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":168,"publishedAt":52},"01M718FW5ST9J56WY4TJ5G5A7X","tres-settings-management","Manage Organization Settings and Platform Settings via the TRES MCP GraphQL API. Use when users ask about org settings, platform settings, configuration, feature flags, enable\u002Fdisable platforms, balance diff, commit strategy, cost basis, ERP, pricing, sync boundaries, or any setting read\u002Fwrite operation. Trigger phrases include \"get settings\", \"show settings\", \"update settings\", \"change settings\", \"enable platform\", \"disable platform\", \"balance diff\", \"commit strategy\", \"cost basis strategy\", \"set min sync date\", \"configure\", \"turn on\", \"turn off\".","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-settings-management\u002FSKILL.md",[167],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[169,170,171,172,149,173,45,44,174,175,176,110,177,178,179,180,181,182,183,184],"tres settings management","settings","management","SET","get","configure","ask","commit","change","organization","update","read","write","sync","mcp graphql","flags",{"skillNo":186,"name":187,"nameZh":6,"description":188,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":189,"categories":190,"keywords":193,"publishedAt":52},"01M718FND35X9DPTNR7JTX2YX8","tres-rollup-rules","Analyze an organization's sub-transaction patterns in TRES Finance, identify groups of repetitive transactions, and propose data-driven rollup rules that consolidate them into one aggregated entry per interval (daily or monthly) for cleaner ERP integration. Also lists and deletes rollup rules. Trigger this skill whenever the user asks about rollup rules, transaction aggregation, consolidating transactions, reducing transaction noise, rolling up gas fees, rolling up staking rewards, rolling up micro-transactions, or collapsing many small ledger entries into fewer summary entries. Also trigger on \"create a rollup rule\", \"suggest rollup rules\", \"what should I roll up\", \"show my rollup rules\", \"delete a rollup rule\", \"aggregate transactions\", \"too many transactions\", \"reduce transaction count\", \"roll up fees\", \"daily rollup\", \"monthly rollup\". If a wallet has too many transactions, high-frequency activity, or noisy ledger entries, proactively suggest rollup rules. For computing the EXACT impact of a single already-specified rule or reviewing submitted\u002Fpending rules, use the tres-rollup-review skill instead.","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-rollup-rules\u002FSKILL.md",[191,192],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":95,"categoryName":96,"taxonomyType":29,"taxonomyTypeDesc":30},[194,195,196,149,197,103,198,45,199,200,201,202,203,154,34,204,205,206,178,207],"tres rollup rules","Rules","rollup","gas","up","impact","reduce","Connector","review","consolidate","transactions","delete","data-driven","activity",{"skillNo":209,"name":210,"nameZh":6,"description":211,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":212,"categories":213,"keywords":216,"publishedAt":52},"01M718FFX3KV2JRFVS2RCM719D","tres-rollup-review","Compute the EXACT sub-transaction impact of a rollup rule in TRES Finance — every filter applied — and flag genuine problems (zero matches, overlap, config mistakes) before the rule is created. Two modes. Mode A: a pre-create quality gate invoked by the tres-rollup-rules skill to vet proposed candidate rules. Mode B: standalone review of a rule someone submitted or of existing PENDING rules. Trigger this skill when the user says \"review this rollup rule\", \"what would this rollup rule actually match\", \"preview the rollup impact\", \"how many transactions would this rule catch\", \"is this rollup rule valid\", \"validate my rollup rule\", \"check my pending rollup rules\", or \"review the rollup requests\". Do NOT use this to design or propose new rollups — that is tres-rollup-rules. This skill only measures impact and flags issues; it never executes mutations.","https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-community\u002Fblob\u002Fmain\u002Ftres-finance-plugin\u002Fskills\u002Ftres-rollup-review\u002FSKILL.md",[214,215],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":95,"categoryName":96,"taxonomyType":29,"taxonomyTypeDesc":30},[202,217,196,218,219,220,221,222,133,223,45,199,44,109,224,201,225,226,34,227],"tres rollup review","standalone","never","mode","issues","design","new","candidate","filter","valid","catch"]