How do I install the Sip Review skill?

Run npx skillmds@latest add getlost01/sip-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Sip Review safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Sip Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Sip Review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Sip Review?

getlost01 (@getlost01) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Sip Review by getlost01 — Agent Skill

Name: Sip Review
Author: getlost01

Sip Review

Health check across all the user's SIPs and funds as a set — expense ratio earned, category-rank drift, manager changes, overlap creep, continue/pause/redirect per SIP. Use when the user asks to review their SIPs, whether to continue one, or for a periodic MF checkup. Read-only — never changes a SIP.

getlost01 8a5ea0c 3.5 KB Updated Sep 17, 2026

File contents

SIP Review

Read-only — never start, stop, pause, or modify a SIP. reference/READ-ONLY-POLICY.md (hard rule) and reference/RESEARCH-STANDARDS.md (MF peer framework, tool availability, data efficiency, disclosure) apply. Where mutual-fund-analysis deep-dives one fund on demand, this sweeps the whole set on a cadence. Works over whichever broker(s) BROKERS.md names active — resolve capabilities against reference/BROKER-CAPABILITIES.md.

Groww's MCP returns no mutual fund data at all. If INDmoney is active, its SIP-status and holdings capability (mf_sips, indian_stocks_sips, networth_holdings — unverified, check the capability map's status table) covers this skill directly; otherwise it runs entirely on PORTFOLIO-PLAN.md plus the web. With no SIP register, no fund list, and no active INDmoney, there is nothing to review — say so and offer portfolio-plan-builder rather than producing a hollow sweep.

Steps

Assemble the set. Funds held, units and cost basis, plus SIP amounts and dates: from INDmoney's holdings/SIP capability if active, otherwise from PORTFOLIO-PLAN.md's holdings and SIP register — no other broker here exposes either. Empty or stale → offer portfolio-plan-builder rather than guessing. Flag any fund listed with no SIP row, and any SIP row with no holding. Note the as-of for every ₹ figure that follows — the register's _Last reviewed:_ date when sourced from the plan, live if from INDmoney.
Per-fund scorecard, peer framework applied lightly across the set — full depth only for funds that flag:
- Rolling returns vs. benchmark and category median across 1Y/3Y/5Y, not one point-to-point window — is the expense ratio earned with actual alpha, net of that expense ratio? Tracking error instead, for index funds. Prefer INDmoney where active; otherwise web-sourced (AMFI / Value Research / factsheet), one date-anchored search per fund at most, cited.
- Category-rank percentile drift across 1Y/3Y/5Y, not a single period: a former top-quartile fund sliding to middling for 2+ years running is a flag; one bad year isn't.
- One date-anchored WebSearch for flagged funds or on a full annual review — manager change, mandate change, AMC issue. fund-house-watch covers AMC-level flags.
Set-level checks: overlap creep (funds converging on the same top holdings — active fees paid repeatedly for one exposure); allocation drift (does SIP flow still match the plan's target mix, or has it skewed, e.g. all new money into small-cap after a hot run); blended expense ratio vs. a passive equivalent.
Verdict per SIP: continue / continue-but-watch / pause / redirect to [bucket], each per the completeness checklist. Redirect targets are buckets; specific fund picks route through mutual-fund-analysis. Any change is the user's action in their broker's app.
Present: one summary table first (fund, SIP ₹, vs. benchmark, rank trend, flag, verdict), then detail only for non-continue verdicts, then set-level flags, then the disclosure block. Formal version: reference/templates/sip-review.md — worth saving for the next review.

---name: sip-reviewdescription: Health check across all the user's SIPs and funds as a set — expense ratio earned, category-rank drift, manager changes, overlap creep, continue/pause/redirect per SIP. Use when the user asks to review their SIPs, whether to continue one, or for a periodic MF checkup. Read-only — never changes a SIP.---# SIP ReviewRead-only — never start, stop, pause, or modify a SIP.`reference/READ-ONLY-POLICY.md` (hard rule) and`reference/RESEARCH-STANDARDS.md` (MF peer framework, toolavailability, data efficiency, disclosure) apply. Where`mutual-fund-analysis` deep-dives one fund on demand, this sweeps thewhole set on a cadence. Works over whichever broker(s) `BROKERS.md`names active — resolve capabilities against`reference/BROKER-CAPABILITIES.md`.Groww's MCP returns no mutual fund data at all. If INDmoney is active,its SIP-status and holdings capability (`mf_sips`, `indian_stocks_sips`,`networth_holdings` — **unverified**, check the capability map's statustable) covers this skill directly; otherwise it runs entirely on`PORTFOLIO-PLAN.md` plus the web. With no SIP register, no fund list, andno active INDmoney, there is nothing to review — say so and offer`portfolio-plan-builder` rather than producing a hollow sweep.## Steps1. **Assemble the set.** Funds held, units and cost basis, plus SIP amounts and dates: from INDmoney's holdings/SIP capability if active, otherwise from `PORTFOLIO-PLAN.md`'s holdings and **SIP register** — no other broker here exposes either. Empty or stale → offer `portfolio-plan-builder` rather than guessing. Flag any fund listed with no SIP row, and any SIP row with no holding. Note the as-of for every ₹ figure that follows — the register's `_Last reviewed:_` date when sourced from the plan, live if from INDmoney.2. **Per-fund scorecard**, peer framework applied lightly across the set — full depth only for funds that flag: - **Rolling** returns vs. benchmark and category median across 1Y/3Y/5Y, not one point-to-point window — is the expense ratio earned with actual alpha, net of that expense ratio? Tracking error instead, for index funds. Prefer INDmoney where active; otherwise web-sourced (AMFI / Value Research / factsheet), one date-anchored search per fund at most, cited. - Category-**rank percentile** drift across 1Y/3Y/5Y, not a single period: a former top-quartile fund sliding to middling for 2+ years running is a flag; one bad year isn't. - One date-anchored `WebSearch` for flagged funds or on a full annual review — manager change, mandate change, AMC issue. `fund-house-watch` covers AMC-level flags.3. **Set-level checks:** overlap creep (funds converging on the same top holdings — active fees paid repeatedly for one exposure); allocation drift (does SIP flow still match the plan's target mix, or has it skewed, e.g. all new money into small-cap after a hot run); blended expense ratio vs. a passive equivalent.4. **Verdict per SIP:** continue / continue-but-watch / pause / redirect to [bucket], each per the completeness checklist. Redirect targets are buckets; specific fund picks route through `mutual-fund-analysis`. Any change is the user's action in their broker's app.5. **Present:** one summary table first (fund, SIP ₹, vs. benchmark, rank trend, flag, verdict), then detail only for non-continue verdicts, then set-level flags, then the disclosure block. Formal version: `reference/templates/sip-review.md` — worth saving for the next review.

Frequently asked questions