agentleFS
grechjgrechj/Australian-Builders-Brain/skills/financials/SKILL.md

financials

Skill0 starsChanged 23 days agoMITView on GitHub

Track job costs and profit for Australian builders. Updates the Budget vs Actual Tracker with committed and spent costs and status, prepares the Job Profit and Loss Summary with gross margin, and runs a lessons-learned review at job close. Use when someone asks how a job is tracking, whether they're over budget, what they've committed or spent, what a job made, their margin or profit, or wants a job wrap-up or post-mortem.

  • Installs packages

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 1 is distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run commands from the user's own project folder (the current working directory), so my_project.yaml and out/ are theirs.

    line 20
  2. Distinctive:

    Before anything else, know who you're helping and where:

    line 42
  3. Very specific:

    Ask for the trade and state if the project file doesn't have them (python3 $ABB/tools/profile.py trades lists every trade.

    line 48
  4. Very specific:

    Write the lines into the tracker (examples/budget_spec.yaml).

    line 65
  5. Very specific:

    Check the margin: python3 $ABB/tools/compute.py margin --param quoted_ex_gst=<total revenue> --param actual_cost_ex_gst=<total costs>.

    line 72
  6. Very specific:

    Write it from the Structure section of brain/documents/<name>.md, which lists every section, field, table and prompt in order.

    line 93
  7. Very specific:

    If they have their own branded Word template, tools/fill_template.py fills any .docx that uses the same [PLACEHOLDERS].

    line 94

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Financials

You help a builder see whether a job is making money while there's still time to act, and learn from it when it's done.

Where things are

The brain (this repository) is at ${CLAUDE_PLUGIN_ROOT}. If that shows literally as ${CLAUDE_PLUGIN_ROOT}, you are working in a clone instead: the brain is the repository root, two levels up from this skill's folder. $ABB below means the brain's folder: write it out as the real path in every command. Paths starting brain/, templates/, tools/ and examples/ are inside it.

Run commands from the user's own project folder (the current working directory), so my_project.yaml and out/ are theirs. Never write into the brain's folder: an installed plugin's folder is replaced on every update.

  • Knowledge: brain/contracts/job_financials.md (read it first), brain/contracts/variations.md, brain/contracts/estimating_and_quoting.md (markup vs margin)
  • Maths: brain/computations/budget_status.md, margin.md, estimate_price.md
  • Templates: templates/excel/Budget_vs_Actual_Tracker_TEMPLATE.xlsx, Job_Profit_Loss_Summary_TEMPLATE.xlsx, Lessons_Learned_Register_TEMPLATE.xlsx
  • Examples: examples/budget_spec.yaml, job_pnl_spec.yaml, lessons_spec.yaml

For python3, use ${CLAUDE_PLUGIN_DATA}/venv/bin/python if it exists (the plugin's setup hook creates it). Otherwise, if python3 -c "import docx, openpyxl, yaml" fails, run python3 -m pip install -r $ABB/requirements.txt first.

To get costs and claims into Xero, MYOB or QuickBooks, for GST, BAS or TPAR figures, or for a dashboard and CSV extracts across jobs (P&L, receipts, costs against budget, margin, time on site, job length against the average), use the accounts skill; its job ledger can feed the actual costs here.

Everything here is ex GST. If the user gives inc-GST invoice totals, convert them with python3 $ABB/tools/compute.py ex_gst --param amount_inc_gst=....

Profile: trade and location

Before anything else, know who you're helping and where:

python3 $ABB/tools/profile.py show --project my_project.yaml      # or: --trade <trade> --state <STATE>

Ask for the trade and state if the project file doesn't have them (python3 $ABB/tools/profile.py trades lists every trade; builder sees everything). Stay within the profile: use the NCC pages, skills, materials and suppliers it lists, apply only that state's NCC variations and rules, and read the state page for local differences.

How's the job tracking?

  1. Get each trade's budget (from the estimate), committed (orders and subcontracts let) and spent (invoices), and whether it's finished.
  2. Work out the status:

    python3 $ABB/tools/compute.py budget_status --param 'lines=[{"item": "Framing", "budget": ..., "committed": ..., "spent": ..., "complete": false}, ...]' \
        > out/budget_receipt.json
    python3 $ABB/brain/references/attesters/check_receipt.py out/budget_receipt.json
  3. Write the lines into the tracker (examples/budget_spec.yaml). The rows are fixed trades (5 Preliminaries to 23 Other; the example lists the row numbers). Columns: B budget, C committed, D spent, H status (from the receipt), I notes.
  4. Tell the user in plain terms: the total forecast variance, which lines are Over Budget or At Risk, and for each the likely cause and fix (variation to recover, back charge, or absorb and record a lesson).

What did the job make?

  1. Collect budget and actual revenue (contract sum, approved variations, extras) and costs by category. Include supervision and the builder's own time.
  2. Fill the Job P&L (examples/job_pnl_spec.yaml): rows 6 to 8 revenue, 11 to 21 costs; B budget, C actual.
  3. Check the margin: python3 $ABB/tools/compute.py margin --param quoted_ex_gst=<total revenue> --param actual_cost_ex_gst=<total costs>. It must match the workbook's Gross Margin. Give both margin and markup, and compare with the margin the job was priced at.

Lessons learned

At job close, ask about each area: estimating, scheduling, subbies, materials, client, safety, quality and commercial. For each lesson, record what happened, the impact, what worked, what to do differently, one action for the next job, and an owner (examples/lessons_spec.yaml). The category must be from the dropdown. Prompt with what the records show: the variation register, delay notices, incidents and budget overruns.

Checks

tools/fill_template.py must report no unmatched entries.

Your own format

The templates are references, not a straitjacket. If the user wants the document on their own letterhead, as an email, as a PDF, or can't run Python (for example in Claude.ai):

  • Write it from the Structure section of brain/documents/<name>.md, which lists every section, field, table and prompt in order. Keep every section, and keep the money exactly as the computation receipt gives it.
  • If they have their own branded Word template, tools/fill_template.py fills any .docx that uses the same [PLACEHOLDERS].

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.