agentleFS
Layneformalized225Layneformalized225/ai-cofounder/skills/fpf-problem-solving/SKILL.md

fpf-problem-solving

Skill1 starsChanged 8 months agoMITView on GitHub

First Principles Framework (FPF) — thinking amplifier. Use when user wants to think through a complex problem, architect a system, evaluate alternatives, decompose complexity, or plan rigorously. Also triggers on: FPF, bounded contexts, SoTA packs, assurance calculus, FPF Parts A-K. Not for simple task planning, general philosophy, or Agile unrelated to FPF.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use FPF whenever you need to think more rigorously than the situation's default.

    line 16
  2. Distinctive:

    Read only the specific sub-section file you need.

    line 57
  3. Distinctive:

    Do NOT load entire sections.

    line 58
  4. Distinctive:

    Use plain language for the user.

    line 62
  5. Distinctive:

    When a problem draws from multiple sections:

    line 67
  6. Distinctive:

    If patterns from different sections appear to conflict, check for category errors via A.7 Strict Distinction — the conflict is usually a …

    line 70
  7. Distinctive:

    Do not just list FPF patterns — weave them into a coherent answer to the user's actual question.

    line 72

≈ “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

First Principles Framework (FPF)

An "Operating System for Thought" — a transdisciplinary architecture for reasoning, written in human- and machine-readable pseudo-code. FPF turns raw intelligence (human or machine) into organisationally usable reasoning: explicit bounded contexts, auditable artefacts, multi-view descriptions, and disciplined hand-offs between specialised actors.

Use cases

Use FPF whenever you need to think more rigorously than the situation's default.

  • Decompose a messy, cross-domain problem into parts that can be reasoned about independently
  • Make a high-stakes decision with incomplete evidence — and know what evidence is still missing
  • Get a mixed team to reason together without vocabulary collisions or hidden assumptions
  • Audit whether a conclusion is well-founded or just plausible
  • Transfer an insight across domains without losing precision or introducing category errors
  • Structure a proposal that must survive scrutiny from multiple expert perspectives
  • Generate alternatives systematically instead of anchoring on the first idea
  • Define what "better" means before comparing options

How to navigate

The use cases above help decide WHETHER to invoke FPF. The router below decides WHERE to go once invoked.

Step 1 — Match the thinking need to a starting point

What you need to doStart here
Decompose a complex whole into bounded parts04 Kernel → A.1 Holons, A.1.1 Bounded Contexts, A.14 Mereology
Assign roles and responsibilities04 Kernel → A.2 Roles, A.15 Role-Method-Work Alignment
Set boundaries on what statements mean05 Signature Stack → classify as definitions, gates, duties, or evidence
Prevent category errors (role vs. function, method vs. work)06 Constitutional Principles → A.7 Strict Distinction
Evaluate confidence in a claim or artifact07 Part B → B.3 Trust & Assurance; 08 Part C → C.2 F-G-R scoring
Compose parts into wholes preserving properties07 Part B → B.1 Gamma algebra; 08 Part C → C.13 Compose-CAL
Reason through a problem systematically07 Part B → B.5 Reasoning Cycle, B.5.2 Abductive Loop
Generate alternatives / explore solution space08 Part C → C.18 NQD Open-Ended Search, C.19 Explore-Exploit
Measure and compare options rigorously06 A.V → A.17-A.19 Characteristics & CSLC; 08 Part C → C.16 MM-CHR
Score knowledge quality (formality, scope, reliability)08 Part C → C.2 KD-CAL, C.2.2 Reliability, C.2.3 Formality
Resolve conflicts across stakeholders or values09 Part D → Ethics & Conflict
Unify vocabulary across teams or domains13 F.I Context of Meaning → 14-15 UTS tables → 20 Lexical Debt
Document for multiple audiences11 E-I Constitution → E.17 Multi-View Publication Kit
Sharpen expression — repair vague wording, surface ambiguity11 E-I → E.17.SD Surface Discipline, E.17.EFP Explanation Faithfulness; 05 A.IV.A → A.6.H Wholeness Unpacking
Survey a discipline and build a reusable toolkit16 Part G → SoTA Packs, TraditionCards, OperatorCards
Trace provenance of a claim06 A.V → A.10 Evidence Graph; 16 Part G → G.6 Provenance Ledger

For complex problems, follow paths across multiple sections — the router shows where to start, not where to stop.

Step 2 — Read the _index.md, then the sub-section

  1. Open the _index.md of the target section folder — it lists all sub-sections with line counts and descriptions.
  2. Read only the specific sub-section file you need.
  3. Do NOT load entire sections. Pick the narrowest file that serves the user's question.

Step 3 — Apply in plain language

Use plain language for the user. Introduce FPF-internal names (U.Holon, Gamma, F-G-R) only when they add precision the user needs.

Step 4 — Compose findings across sections

When a problem draws from multiple sections:

  1. State each pattern's contribution in one line (e.g., "Bounded Contexts gives us the parts; Trust Calculus scores our confidence in each").
  2. If patterns from different sections appear to conflict, check for category errors via A.7 Strict Distinction — the conflict is usually a level confusion (role vs. function, method vs. work), not a real contradiction.
  3. Synthesize in natural order: decomposition first (what are the parts?), then evaluation (how confident are we?), then resolution (what do we do about gaps?).
  4. Do not just list FPF patterns — weave them into a coherent answer to the user's actual question.

Starter prompt (example — adapt to the user's actual role and need)

You have the FPF specification loaded. Help me structure my project / problem / programme. Use plain language for an engineer-manager. Propose: (1) bounded contexts / specialisations, (2) decision criteria, (3) key alternatives, (4) hand-offs, and (5) missing evidence or tests before commitment. Introduce internal FPF names only when they add precision.

Section INDEX

Structural reference. Each entry is a folder — read its _index.md first, then pick the sub-section.

#SectionSubWhen to use
01Title page0Authorship, version date, top-level identity.
02Table of Content0Navigate the spec, locate a pattern, trace inter-section dependencies.
03Preface17Onboard: reading paths by role, FPF philosophy, purpose and non-goals.
04Part A — Kernel19Decompose and assign: holons, bounded contexts, roles, transformers, method/work separation.
05A.IV.A — Signatures20Set boundaries: classify statements as definitions, gates, duties, or evidence.
06A.V — Principles29Prevent confusion: category errors, measuring, comparing, evidence graphs.
07Part B — Reasoning24Compose and evaluate: aggregation (Gamma), trust scores, emergence, reasoning cycles.
08Part C — Extensions30Score and search: epistemic quality (F-G-R), kinds, measurement, open-ended search.
09Part D — Ethics1Resolve conflicts: ethical trade-offs, bias auditing, safety overrides.
10Part E — Constitution0Entry point for Part E subsections.
11E-I — Constitution33Govern and publish: 11 Pillars, guard-rails, multi-view publication (MVPK), surface discipline.
12Part F — Unification0Entry point for Part F subsections.
13F.I — Meaning19Align vocabulary: semantic drift, homonym collisions, Alignment Bridges.
14UTS Layout A0Map concepts across standards (BPMN, PROV-O, ITIL).
15UTS Layout B1Map concepts across disciplines (operations, physics, math).
16Part G — SoTA Kit15Harvest disciplines: SoTA Packs, TraditionCards, OperatorCards, benchmarks.
17Part H — Glossary0Look up terms: canonical definitions, four-register naming, cross-references.
18Part I — Annexes0Walkthroughs, change log, external standards mappings.
19Part J — Indexes0Concept-to-pattern, pattern-to-example, principle-trace indexes.
20Part K — Lexical Debt2Fix terminology: mandatory replacements and migration debt.

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.