--- name: pp-fpi-india description: "The full history of Foreign Portfolio Investor flows into India, synced locally and queryable offline — something neither NSDL nor CDSL offers today. Trigger phrases: `FPI flows into India`, `foreign portfolio investor data`, `NSDL FPI report`, `FII DII flow India`, `check FPI net investment`, `use fpi-india`, `run fpi-india`." author: "Mayank Lavania" license: "Apache-2.0" argument-hint: " [args] | install cli|mcp" allowed-tools: "Read Bash" metadata: openclaw: requires: bins: - fpi-india-pp-cli install: - kind: go bins: [fpi-india-pp-cli] module: github.com/mvanhorn/printing-press-library/library/other/fpi-india/cmd/fpi-india-pp-cli --- # FPI India — Printing Press CLI ## Prerequisites: Install the CLI This skill drives the `fpi-india-pp-cli` binary. **You must verify the CLI is installed before invoking any command from this skill.** If it is missing, install it first: 1. Install via the Printing Press installer. It defaults binaries to `$HOME/.local/bin` on macOS/Linux and `%LOCALAPPDATA%\Programs\PrintingPress\bin` on Windows: ```bash npx -y @mvanhorn/printing-press-library install fpi-india --cli-only ``` 2. Verify: `fpi-india-pp-cli --version` 3. Ensure the reported install directory is on `$PATH` for the agent/runtime that will invoke this skill. If the `npx` install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.6 or newer). This installs into `$GOPATH/bin` (default `$HOME/go/bin`), so add that directory to `$PATH` instead: ```bash go install github.com/mvanhorn/printing-press-library/library/other/fpi-india/cmd/fpi-india-pp-cli@latest ``` If `--version` reports "command not found" after install, the runtime cannot see the binary directory on `$PATH`. Do not proceed with skill commands until verification succeeds. NSDL and CDSL publish authoritative FPI investment data, but only as ASP.NET report pages with a client-side Excel export button — no API, no bulk historical download, no scriptable access. fpi-india syncs the full net-investment series back to 1992-93, plus AUC, sector rotation, trade data, and the FPI registry, into a local SQLite database you can query, trend, and pipe. ## When to Use This CLI Use this CLI for historical or current Indian FPI investment flow analysis: net-investment trends, year-over-year and streak analysis, sector rotation, assets-under-custody trends, FPI registry lookups, and regulatory limit utilization. It is the right tool whenever a task needs NSDL/CDSL FPI data queried, trended, or compared rather than eyeballed from a report page. ## Anti-triggers Do not use this CLI for: - Do not use this CLI for live/real-time trade execution or order placement — it is a read-only data and analytics tool. - Do not use this CLI for individual investor portfolio holdings or demat account data — NSDL/CDSL's FPI reports are aggregate market-level data, not account-level. - Do not use this CLI for forward-looking flow forecasts or investment advice — every command surfaces historical, already-published figures only. ## Unique Capabilities These capabilities aren't available in any other tool for this API. ### Local state that compounds - **`net-investment yoy`** — See this year's FPI net flow next to the same period last year, delta computed automatically. _Reach for this when a request compares FPI flow across two adjacent years instead of listing raw rows._ ```bash fpi-india-pp-cli net-investment yoy --asset equity --year 2024 --json ``` - **`sector rotation`** — Rank which sectors saw the biggest FPI inflow/outflow swing between the two most recent synced periods. _Use this for 'what sector are FPIs rotating into/out of' instead of a single-period sector snapshot._ ```bash fpi-india-pp-cli sector rotation --top 5 --period latest --json ``` - **`limits utilization`** — See what percentage of the regulatory FPI investment cap a sector or debt category has used. _Use this to flag sectors approaching a regulatory FPI cap, which can trigger a trading halt._ ```bash fpi-india-pp-cli limits utilization --sector Banking --json ``` - **`net-investment streaks`** — See how many consecutive fortnights/years FPIs have been net buyers or sellers, and the most recent flip. _Use this for 'how many periods in a row' questions instead of scanning raw historical rows._ ```bash fpi-india-pp-cli net-investment streaks --asset equity --json ``` - **`auc trend`** — Track how a country's or category's share of assets under custody has changed across synced snapshots. _Use this once at least two snapshots are synced to see custody-share direction, not a single-point value._ ```bash fpi-india-pp-cli auc trend --by country --country Mauritius --json ``` - **`net-investment extremes`** — Find the largest single-period FPI inflows or outflows across the full 1992-93-to-present series. _Use this for 'biggest ever' / superlative questions about historical FPI flow._ ```bash fpi-india-pp-cli net-investment extremes --asset equity --top 10 --json ``` - **`net-investment trend`** — See the rolling growth-rate and direction of FPI net flow over a window of recent periods. _Use this for a derived growth-rate/direction view, not raw historical rows (use 'net-investment fy'/'cy'/'quarterly'/'latest' for those)._ ```bash fpi-india-pp-cli net-investment trend --asset equity --period fy --window 12 --json ``` ### Reachability mitigation - **`cdsl diff`** — Compare NSDL's and CDSL's figures for the same date and flag any mismatch. _Use this to cross-check the two Indian depositories' published figures for the same date._ ```bash fpi-india-pp-cli cdsl diff --date 09072026 --json ``` ## Command Reference **auc** — FPI assets under custody (AUC), by country and by category - `fpi-india-pp-cli auc category` — FPI AUC category-wise data - `fpi-india-pp-cli auc country` — FPI AUC country-wise data (top 10 countries) **cdsl_reports** — CDSL daily FPI snapshot publication index and dated XLS snapshot download - `fpi-india-pp-cli cdsl-reports listing` — Index page listing recent dated FPI snapshot XLS files - `fpi-india-pp-cli cdsl-reports snapshot` — Dated daily FPI snapshot file (date format DDMMYYYY, e.g. 09072026) **limits** — Per-company (ISIN) foreign investment limits: NRI limit %, FPI limit %, sectoral cap %, and current monitored FPI holding count - `fpi-india-pp-cli limits` — Per-ISIN foreign investment limit monitoring data **net_investment** — Historical FPI net-investment flows by asset class (equity, debt, hybrid, mutual funds, AIF), financial year and calendar year - `fpi-india-pp-cli net-investment cy` — Calendar-year FPI net-investment series, with monthly breakdown when a specific year is requested - `fpi-india-pp-cli net-investment fy` — Full financial-year FPI net-investment series (1992-93 to present) - `fpi-india-pp-cli net-investment latest` — Latest fortnight FPI net-investment snapshot - `fpi-india-pp-cli net-investment quarterly` — Quarterly FPI net-investment details (calendar year) **registry** — Registered FPI list, registration category counts, and DDP pendency - `fpi-india-pp-cli registry categories` — FPI category-wise registration counts - `fpi-india-pp-cli registry list` — List of registered FPIs/FIIs - `fpi-india-pp-cli registry pendency` — DDP-wise pendency of FPI applications **sector** — Fortnightly sector-wise FPI investment data - `fpi-india-pp-cli sector` — Fortnightly sector-wise FPI investment activity **trades** — Trade-wise FPI equity and debt data - `fpi-india-pp-cli trades debt` — Trade-wise debt data of FPI - `fpi-india-pp-cli trades equity` — Trade-wise equity data of FPI ### Finding the right command When you know what you want to do but not which command does it, ask the CLI directly: ```bash fpi-india-pp-cli which "" ``` `which` resolves a natural-language capability query to the best matching command from this CLI's curated feature index. Exit code `0` means at least one match; exit code `2` means no confident match — fall back to `--help` or use a narrower query. ## Recipes ### Full historical equity flow series ```bash fpi-india-pp-cli net-investment fy --json --select period,total,cumulative_total ``` Narrow a 30+ year series down to just the fields needed for a chart or note. ### This fortnight vs last year, same fortnight ```bash fpi-india-pp-cli net-investment yoy --asset equity --year 2024 ``` The Monday-morning commentary workflow: one command instead of loading two pages and subtracting by hand. ### What sector is money rotating into ```bash fpi-india-pp-cli sector rotation --top 5 --period latest ``` Ranks period-over-period sector deltas instead of a single period's absolute figures. ### Is any sector near its FPI investment cap ```bash fpi-india-pp-cli limits utilization --sector Banking ``` Joins synced limits against synced investment totals to compute percent-of-cap used. ### Biggest FPI outflow ever recorded ```bash fpi-india-pp-cli net-investment extremes --asset equity --top 10 ``` Deterministic ranking over the full synced history for journalist-style superlative questions. ## Auth Setup No authentication required. Run `fpi-india-pp-cli doctor` to verify setup. ## Agent Mode Add `--agent` to any command. Expands to: `--json --compact --no-input --no-color --yes`. - **Pipeable** — JSON on stdout, errors on stderr - **Filterable** — `--select` keeps a subset of fields. Dotted paths descend into nested structures; arrays traverse element-wise. Critical for keeping context small on verbose APIs: ```bash fpi-india-pp-cli limits --action GetFilmReport_Detail --u-name example-resource --agent --select id,name,status ``` - **Previewable** — `--dry-run` shows the request without sending - **Offline-friendly** — sync/search commands can use the local SQLite store when available - **Non-interactive** — never prompts, every input is a flag - **Read-only** — do not use this CLI for create, update, delete, publish, comment, upvote, invite, order, send, or other mutating requests ### Response envelope Commands that read from the local store or the API wrap output in a provenance envelope: ```json { "meta": {"source": "live" | "local", "synced_at": "...", "reason": "..."}, "results": } ``` Parse `.results` for data and `.meta.source` to know whether it's live or local. A human-readable `N results (live)` summary is printed to stderr only when stdout is a terminal AND no machine-format flag (`--json`, `--csv`, `--compact`, `--quiet`, `--plain`, `--select`) is set — piped/agent consumers and explicit-format runs get pure JSON on stdout. ## Paths and state Agents should treat the CLI's path resolver as part of the runtime contract: - Use `--home ` for one invocation, or set `FPI_INDIA_HOME=` to relocate all four path kinds under one root. - Use per-kind env vars only when a specific kind must diverge: `FPI_INDIA_CONFIG_DIR`, `FPI_INDIA_DATA_DIR`, `FPI_INDIA_STATE_DIR`, `FPI_INDIA_CACHE_DIR`. - Resolution order is per-kind env var, `--home`, `FPI_INDIA_HOME`, XDG (`XDG_CONFIG_HOME`, `XDG_DATA_HOME`, `XDG_STATE_HOME`, `XDG_CACHE_HOME`), then platform defaults. - `config` contains settings like `config.toml` and profiles. `data` contains `credentials.toml`, `data.db`, cookies, and auth sidecars. `state` contains persisted queries, jobs, and `teach.log`. `cache` contains regenerable HTTP/cache files. - Stored secrets live in `credentials.toml` under the data dir. Existing legacy `config.toml` secrets are read for compatibility and leave `config.toml` on the first auth write. - Run `fpi-india-pp-cli doctor --fail-on warn` to surface path and credential-location warnings. `agent-context` exposes a schema v4 `paths` block for agents that need the resolved dirs. - For MCP, pass relocation through the MCP host config. The MCP binary does not inherit CLI flags: ```json { "mcpServers": { "fpi-india": { "command": "fpi-india-pp-mcp", "env": { "FPI_INDIA_HOME": "/srv/fpi-india" } } } } ``` Fleet precedence: an inherited per-kind env var overrides an explicit `--home` for that kind. Use `FPI_INDIA_HOME` or per-kind vars as durable fleet levers, and use `--home` only for a single invocation. Relocation is not reversible by unsetting env vars; move files manually before clearing `FPI_INDIA_HOME`, or `doctor` will not find credentials left under the former root. ## Automatic learning This CLI ships a self-capturing learning loop. The CLI does its own bookkeeping: every invocation is journaled locally, a failed flag followed by a corrected retry auto-derives a `flag_alias` candidate, and a `teach` on a query family without a playbook auto-synthesizes a `playbook_candidate` from the session's journal. Your job is judgment only: `recall` first, act on surfaced candidates, `teach` the final answer, `playbook amend` when you observe a correction. You never record failures by hand. ### Step 1: `recall` before any discovery Before list/search/drill commands on a new user question, run: ```bash fpi-india-pp-cli recall "" --agent ``` The response envelope: ```json { "query": "...", "normalized": "", "query_entities": ["..."], "found": true | false, "match_score": 0.0, "results": [ { "resource_id": "...", "resource_type": "...", "venue": "...", "confidence": 2, "entity_match": "exact|partial|unknown", "source": "taught|preseed|pattern", "warnings": ["..."] } ], "mismatches": [ /* only when --debug-mismatches */ ], "warnings": [ /* top-level */ ], "candidates": [ { "id": 12, "class": "flag_alias | playbook_candidate", "summary": "...", "sightings": 3, "last_seen": "...", "rationale": "...", "next_action": ["", "fpi-india-pp-cli learnings confirm 12"] } ], "playbook": { "query_family": "...", "playbook": { "steps": [ { "cmd": "", "purpose": "..." } ], "entity_slots": ["$ENTITY"], "expected_tool_calls": 3 }, "slots_resolved": { "$ENTITY": { "token": "", "canonical": "" } }, "notes": "" }, "notes": "" } ``` Empty-store short-circuit: if the store has no learnings, playbooks, or candidates yet (recall finds nothing and `learnings list` and `learnings candidates` are both empty), skip recall for the rest of this session instead of taxing every query; resume recall-first once something has been taught. ### Step 2: decision tree Read `candidates`, `playbook`, `notes`, `results[0]`, and warnings in that order: ``` if Candidates present (warnings include "candidates_present"): -> candidates are try-then-confirm, never facts. Follow each candidate's two-step next_action verbatim: run the trial command first, then run `learnings confirm ` only after the trial verified the behavior. Reject a wrong candidate with `learnings reject `. -> NEVER re-teach something recall surfaced as a candidate; confirm or reject that candidate instead of teaching a duplicate. -> candidates ride alongside playbooks and resource hits, not instead of them; continue with the branches below after acting on them. if Playbook present: -> READ Playbook.notes verbatim FIRST (workarounds + gotchas the CLI surface doesn't expose) -> replay Playbook.steps in order, substituting Playbook.slots_resolved entries for the entity slot tokens. If a step's slot is unresolved, fall back to discovery for that step only. -> the Playbook's expected_tool_calls is a budget; if you find yourself running materially more, record the divergence via `fpi-india-pp-cli playbook amend` at end-of-session. elif Notes present (no Playbook): -> read Notes verbatim before any discovery step; they carry known gotchas for this query family even when no structured choreography exists yet. elif Found AND Results[0].EntityMatch == "exact" AND Results[0].Confidence >= 2: -> skip discovery; fetch live data for Results[*].ResourceID in parallel elif Found AND Results[0].EntityMatch == "partial": -> candidate hint, NOT a hit; read the resource title to validate before trusting elif (any row in Mismatches[] when --debug-mismatches was passed): -> treat as cold start; the stored learning is for a different entity (different canonical resolved from query_entities) else: // Found == false, no playbook, no notes -> cold start; run discovery normally; teach the answer afterward (Step 4). If the family has no playbook yet, that teach auto-synthesizes a playbook candidate from this session's journal - you do not need to record one by hand. ``` Playbook and Notes are orthogonal to the per-resource path. A recall response can carry both a Playbook AND a `Results[]` hit - use both: the Playbook tells you which choreography to run; the resource hits short-circuit specific steps. Default to skipping `mismatches`; pass `--debug-mismatches` only when investigating cold-start surprises. Candidate judgment details: `learnings confirm ` prints the candidate's full payload before materializing it - check that the printed payload matches the behavior you verified. `learnings reject ` tombstones the derivation signature so the same candidate does not resurface. The envelope carries only the few candidates worth acting on now; `fpi-india-pp-cli learnings candidates` lists the full open set. Graceful degradation: if `learnings confirm` is an unknown command, you are driving an older binary - ignore the candidates guidance and follow the rest of the protocol. ### Step 3: always read `warnings` - `low_confidence`: row exists at `confidence<2`. Treat as a hint, not a skip-discovery hit. - `resource_not_in_store`: the local store doesn't have the resource the learning points at. The match validator couldn't classify entities — direct-fetch and re-evaluate. - `cross_alias_match` (per-result): the row was taught under a different alias and matched the live query's canonical via `entity_lookups` (e.g., a "USA" teach satisfying a "United States" recall). Trust the resource_id. - `similar_shape_different_entity:` (top-level): a structurally matching row exists but its canonical entity differs from the live query's. Treated as cold start; the warning carries the conflicting canonical as a hint, but the row is NOT promoted into Results. - `ambiguous_alias` (top-level): a single query entity resolved to multiple canonicals (e.g., "Cards" → Arizona Cardinals + St. Louis Cardinals). Surface the ambiguity from context before committing to a resource. - `candidates_present` (top-level): the envelope carries a `candidates` section. Handle it via the candidates branch in Step 2 before anything else. - `lookup_refresh_available` (top-level): an entity in the query has no lookup row yet, but synced data could provide one. Run `fpi-india-pp-cli sync` to refresh entity lookups. - Top-level `no_learnings_for_query_family`: the table had no rows above the Jaccard floor. Pure cold start. ### Step 4: `teach &` after finalizing your response - always Teaching is unconditional. After resolving a query the store could not answer, background-teach the final resource mapping - no call-count threshold, n