agentleFS
tonytony/skills/.agents/skills/business-report-leadership/SKILL.md

business-report-leadership

Skill0 starsChanged 17 days agoMITView on GitHub

Use when rendering a tier-0 leadership report from a completed business-value run, with SCQA and a conservative value case.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Read references/interim-format.md first.

    line 18
  2. Very specific:

    Run references/business.py breakeven <run> (absolute path).

    line 41
  3. Very specific:

    Write <run>/reports/leadership.md with this fixed structure:

    line 56
  4. Very specific:

    Write it per references/prose-rules.md.

    line 84

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Leadership Report (tier 0)

Render the full-detail leadership report from a run package. Tier 0 withholds nothing: names, raw links, registers inline.

Read references/interim-format.md first. It defines the run location, the completeness gate, and the output frame. Defer these until the gate passes:

  • references/provenance.md
  • references/measurement.md: the value build, break-even, guardrails, the counterfactual ladder.
  • references/evidence.md: external context for the pre-answered objections. Re-verify the sources you cite first, per its § Re-verify before citing; do not narrate a missing fetch tool.
  • references/audiences.md: the tier-0 contract.

User arguments: $ARGUMENTS

Procedure

1. Locate the run and pass the gate

Per interim-format.md § Rendering a report; a request that names a workflow resolves per its § Locating an existing run.

2. Compute the figures

Run references/business.py breakeven <run> (absolute path). For an interval, run references/business.py stats --csv <file> --column <column> on a raw/ CSV, or pass one number per line on stdin. Every computed figure in the report (S_inst, S_net, N_breakeven, and the conservative, base, and upside scenarios) comes from that output, copied as printed; never compute, round, or convert one yourself. The conservative lines are the committed figures. Each missing: line is an unknown naming that register id, with the data that would resolve it. Printed problem lines are a register error: fix the register first. Exit 1 with missing: lines only means nothing is computable: render an evidence report per measurement.md, with the measured medians only and break-even stated as unknown.

3. Render the report

Write <run>/reports/leadership.md with this fixed structure:

  1. SCQA opening: Situation, Complication, Question, Answer, with the Answer stated first.
  2. Exhibits: every exhibit title is a full-sentence takeaway; one message per exhibit; a source, query, and date-range note under each.
  3. Value build: every multiplier explicit, per the chains in measurement.md; break-even from the script output, or stated unknown.
  4. Scenarios: the conservative scenario (defined in measurement.md) is the committed number; base and upside appear as context, never as the headline. With negative per-instance savings, state the actual volume bounds used: high for conservative, low for upside.
  5. Assumptions register and data dictionary: appendix, inline from assumptions.yaml plus definitions of every metric used.
  6. Known limitations: each with a confidence label.
  7. What would flip this conclusion: the input thresholds at which the answer changes, from registered or scripted figures; otherwise name the data needed, never an invented trial-count cutoff. Recommend no sample count without a registered precision or power design; display thresholds are not collection targets.
  8. Pre-answered objections: at minimum: how do you know the skill caused this (name the counterfactual rung); does the savings rate survive the verification tax; who maintains the skill. The maintainer is a run-package fact only when the register or findings name an owner; otherwise ask the user once, or write "unknown" with the data that would resolve it.

Write it per references/prose-rules.md.

Then run references/business.py trace <run>/reports/leadership.md --run <run> (absolute path). Each number it prints is absent from the breakeven and stats output and from the run's records: replace it with the printed figure and trace again until ok. Then run steps 4 and 5 of interim-format.md § Rendering a report: the money check, then stamp.

Output

Hero success form: ✓ Leadership report: net <signed conservative figure> engineer-hours projected over <horizon>, upside <signed figure> (<tag>); keep the printed sign, never write "loses" before a negative number. When the script prints no upside, name the unknown input instead. Sections and stopped-run shape as in interim-format.md; ## Report also names the strongest limitation. Panel options: render the org-wide projection next (read the business-report-org-wide skill and follow it), adjust assumptions and re-render, or stop.

Portability notes

  • $ARGUMENTS — the text the user passed when invoking this skill. If your host does not substitute it, read it as the user's request in the current turn, and ask when there is none.
  • Bundled files — every relative path in this skill points at a file shipped inside this skill directory. Read them from here, not from the host's plugin tree.

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.