fin-personal-cfo
Skill0 starsChanged 46 days agoMITView on GitHub
Model household financial choices and compare scenarios with disclosed inputs and sensitivity. Use for "house math", "rental forecast", "renovation budget", "monthly cash flow", "big purchase", or "savings what-if". Includes a detailed housing template and general scenario methods. Company books belong in fin-company-cfo; this skill provides scenario analysis and never executes purchases, loans, investments, or tax filings.
In this repository96 other files it gives its agents.
- AGENTS.mdAGENTS.md
- team_context/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- fin-company-cfoSkill
- fin-invoice-reconciliationSkill
- fin-month-end-reportingSkill
- meta-find-skillsSkill
- meta-memory-recallSkill
- meta-memory-writeSkill
- meta-skill-creatorSkill
- meta-skillifySkill
- meta-synthesize-localsSkill
- meta-toolifySkill
- meta-wrap-upSkill
- mkt-ab-testingSkill
- mkt-ad-creativeSkill
- mkt-adsSkill
- mkt-asoSkill
- mkt-brand-voiceSkill
- mkt-churn-preventionSkill
- mkt-cold-emailSkill
- mkt-co-marketingSkill
- mkt-community-marketingSkill
- mkt-content-repurposingSkill
- mkt-copy-editingSkill
- mkt-copywritingSkill
- mkt-croSkill
- mkt-directory-submissionsSkill
- mkt-emailsSkill
- mkt-eventsSkill
- mkt-free-toolsSkill
- mkt-icpSkill
- mkt-influencer-marketingSkill
- mkt-jab-hookSkill
- mkt-launchSkill
- mkt-lead-magnetsSkill
- mkt-marketing-loopsSkill
- mkt-offersSkill
- mkt-onboardingSkill
- mkt-paywallsSkill
- mkt-popupsSkill
- mkt-positioningSkill
- mkt-programmatic-seoSkill
- mkt-prospectingSkill
- mkt-public-relationsSkill
- mkt-referralsSkill
- mkt-sales-enablementSkill
- mkt-schemaSkill
- mkt-signupSkill
- mkt-smsSkill
- mkt-socialSkill
- mkt-ugc-scriptsSkill
- mkt-visual-identitySkill
- ops-analyticsSkill
- ops-attributionSkill
- ops-cronSkill
- ops-ingestSkill
- ops-loopifySkill
- ops-project-managementSkill
- ops-revopsSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
17 rules. About 0 are common, 8 are distinctive, and 9 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 36
Use the linked skills only if installed.
- Very specific: line 38
Keep durable source pointers with the dated project report.
- Very specific: line 44
Read relevant assets/ examples if present.
- Very specific: line 48
Use references/templates/house.md for housing.
- Distinctive: line 49
Read the detailed method for this step’s mechanics and output shape.
- Very specific: line 58
Read references/calculators.md.
- Distinctive: line 63
Write comparison, assumption ledger, detailed report, and unresolved questions.
- Very specific: line 68
Create projects/fin-personal-cfo/{YYYY-MM-DD}_{name}/ in the selected workspace.
- Very specific: line 70
Verify factual/source coverage, schema, calculations, destination fit, and links as appropriate.
- Very specific: line 76
Ask for feedback after major deliverables.
- Distinctive: line 82
Read Rules and local overrides before every Solo run.
- Distinctive: line 83
Check actual connector/CLI availability and authentication.
- Distinctive: line 84
Never send, post, schedule publication, purchase, or push to a remote merely because a source example shows it.
- Distinctive: line 85
Keep base definitions immutable during ordinary use.
- Distinctive: line 86
No financial transactions or filings.
- Distinctive: line 87
Use current sourced evidence or clearly disclosed assumptions, never fabricated rates or comps.
- Very specific: line 92
Create the local override if absent.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Personal CFO Scenarios
An AI-OS capability with scoped context, local configuration, and reviewable outputs.
Outcome
Dated input data, auditable calculations, a scenario comparison, and sensitivity report.
Save deliverables to projects/fin-personal-cfo/{YYYY-MM-DD}_{name}/ with dated filenames.
Knowledge skills also maintain their scoped raw/wiki corpus; integration and authoring skills may modify the explicitly selected project.
Context Needs
| File | Load level | Purpose |
context/learnings.md | ## fin-personal-cfo only | Feedback for this skill |
context/learnings.md | Relevant General entries | Cross-skill corrections |
context/config/fin-personal-cfo/ | Relevant files if present | User-owned settings |
Adjacent SKILL.local.md | Full if present | User overrides; take precedence |
Dependencies
| Skill | Required? | What it provides | Without it |
str-deep-research | Optional | Verify current rates, rents, and comparable evidence | Use disclosed user-supplied assumptions |
str-decide | Optional | Record the eventual choice | Present cash/lifestyle tradeoffs |
tool-humanizer | Optional for publishable prose | Editorial gate | Natural-language, voice, factual-claim checklist |
Skill Relationships
Use the linked skills only if installed; check their SKILL.md and local overrides before invoking.
Consume related source artifacts without duplicating their workflow. Keep trigger boundaries stated in the description.
Keep durable source pointers with the dated project report. Use meta-memory-write only when the user asks to remember a concise working fact; full research and financial records stay in project files.
Step 1: Load scope and corrections
Check the runtime before reading local context. With Team connected, the authoritative workspace snapshot supplies context and feedback; do not read or write local brand_context, context, or learnings. Use its configured output/feedback mechanisms and report a missing snapshot instead of a local fallback. In Solo mode, resolve the AI-OS root or explicitly active client and read only that scope’s listed brand files, this skill’s learnings, relevant General feedback, and local override.
In Solo mode, resolve user configuration as ${AI_OS_SKILL_CONFIG_DIR:-context/config}/fin-personal-cfo/ relative to that workspace; do not reuse another client’s settings.
Read relevant assets/ examples if present; use the output template until an approved example exists.
Step 2: Select scenario and gather inputs
Use references/templates/house.md for housing. Other modes—refi, cash flow, big purchase, side income, savings—use the general calculator methods and a task-specific schema, not nonexistent templates. Establish jurisdiction, currency, horizon, income, liabilities, costs, and provenance for each input. Read the detailed method for this step’s mechanics and output shape.
Step 3: Build realistic comparisons
For housing compare baseline, accessory unit, bedroom rentals, combinations, and staged rentals when relevant. Include closing cash, renovations, vacancy, owner-paid utilities, fees, taxes, insurance, and personal comfort. Let the user’s property and lifestyle determine the actual scenario set. Read the detailed method for this step’s mechanics and output shape.
Step 4: Calculate and audit
Read references/calculators.md. Compute debt service, monthly/annual net costs, initial cash, paydown, and alternative-return scenarios. Use consistent units and distinguish nominal versus real returns. Tax effects are jurisdiction-dependent estimates, not automatic US deductions. Test vacancy, rent, rates, maintenance, renovation overruns, and returns. Read the detailed method for this step’s mechanics and output shape.
Step 5: Present sensitivity and decision inputs
Write comparison, assumption ledger, detailed report, and unresolved questions. Show which choice is cheapest, which is resilient, and how lifestyle changes the answer. Highlight thresholds and downside scenarios rather than a single optimistic point estimate; do not treat modeled returns as guaranteed. Read the detailed method for this step’s mechanics and output shape.
Step 6: Save and verify output
Create projects/fin-personal-cfo/{YYYY-MM-DD}_{name}/ in the selected workspace.
Date-stamp generated filenames as {YYYY-MM-DD}_{descriptive-name}.{ext}. Keep an index of prior deliverables when revisits are useful.
Verify factual/source coverage, schema, calculations, destination fit, and links as appropriate. Store raw inputs only in the authorized local scope; do not mix client data.
For publishable prose only, run tool-humanizer if installed; otherwise edit for natural language, concrete claims, the user’s voice, and unsupported promises. Do not rewrite quotations, code, or structured data through this gate.
Show a concise preview, report the full absolute deliverable path, and distinguish completed work from unmet dependencies.
Step 7: Capture feedback
Ask for feedback after major deliverables. In Solo mode append dated, contextual feedback under ## fin-personal-cfo in the scoped context/learnings.md.
Save an approved good output as an asset only when the user authorizes it; keep private source data out of shared assets.
Rules
- Team snapshot authority takes precedence over all local context/config/knowledge instructions in this entrypoint and its references.
- Read Rules and local overrides before every Solo run. User task authorization and AI-OS project rules govern actions.
- Check actual connector/CLI availability and authentication; use a documented local/manual fallback when absent.
- Never send, post, schedule publication, purchase, or push to a remote merely because a source example shows it.
- Keep base definitions immutable during ordinary use; store corrections in SKILL.local.md.
- No financial transactions or filings; assumptions and uncertainty are part of the deliverable.
- Use current sourced evidence or clearly disclosed assumptions, never fabricated rates or comps.
Self-Update
With Team connected, use only the authoritative snapshot’s configured feedback/update mechanism; do not modify local context or skill rules. In Solo mode, when the user flags a wrong approach, format, assumption, or tone, immediately add the dated correction to ## Rules in adjacent SKILL.local.md, then log it under ## fin-personal-cfo in context/learnings.md.
Create the local override if absent. Preserve existing local entries and leave the shipped SKILL.md unchanged.
Troubleshooting
- Missing configuration or source: ask for the necessary input, then continue independent local work.
- Missing optional tool: use the declared fallback and state the coverage limit.
- Missing required native skill: stop only the dependent step and report its exact name.
- Reference command differs from installed tooling: read current official docs; do not pretend the command succeeded.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

