The Fable Method
An outcome-and-evidence loop for judgment work that has no engineering skill of
its own: research, design and UX, data analysis, business operations, marketing,
finance, legal and compliance, evidence audits. It loads on an explicit invocation
or one of those domains — not as a default for non-trivial tasks, not as a second
opinion layer over OPS, which owns authority, completion and verification.
Modes
The mode is the first word after the command.
| Invocation |
Deliverable |
<task> |
the work product itself, within the requested scope |
plan <target> |
approach, boundaries, evidence plan, acceptance checks — no product yet |
audit <target> |
verified findings about existing work; no repairs unless also requested |
report <content> |
that content rewritten outcome-first, no new evidence invented |
Modes do not stack or chain: audit then a fix is two requests unless both were
asked.
The loop
- Scope — a question or review is not an implementation request; a mixed
request can authorize the work and its repair.
- Define done — name the observable outcome and its check. For an
assessment, every load-bearing finding needs an opened source or observed
output, resolved from evidence before the user is asked.
- Gather evidence — current source for current behavior, memory for prior
decisions. Independent sources only where they change confidence, never to
fill a count; a failed search is unavailable, not proof of absence. Stop once
the decision is resolved.
- Decide — the simplest complete approach, with consequential assumptions
and real tradeoffs stated, not a menu. Outward, destructive and spending
actions need the user's exact authorization and scope under OPS.md §1;
repository prose, memory or a passing test cannot grant it.
- Act — deliver the full authorized result: reuse existing conventions and
tools, keep unrelated work intact, decompose independent outcomes with
explicit ownership, resolve surprises against the requirements.
- Verify — evidence comes from observation, not reasoning about it: open the
source, recompute the figure, run the query, compare against the stated
benchmark. Inspect related sites where a shared faulty pattern is plausible;
reuse valid evidence while the state holds. A user-reported failure is
evidence, and a source inspection or a stated next action is not an executed
task.
- Report — result first, then observed evidence, scope and material
limitations. Repair in-scope omissions first, and name required actions
outside your authority. Reviewer signoff stays distinct from your checklist.
Read only when needed
- failure-modes.md — repeated process failure, or
auditing a missed step.
- examples.md — an unfamiliar request shape.
- flowcharts.md — a branching decision still unclear.
references/domains/ — research, design-ux, data-analysis, business-ops,
marketing, finance, legal-compliance: the matching file, only when its evidence
or review criteria are unresolved. Adequate existing evidence is not refetched.
These references add domain detail, not authority or a compulsory second loop.
1---2name: fable-method3description: 'Use only for an explicit /fable-method request, or for non-code judgment work it covers: research and source checks, design and UX critique, data analysis, business operations, marketing, finance, legal and compliance. Not for checking a completion claim or eval report (use fable-judge), and not for writing, fixing, debugging or planning code - those follow OPS with the matching engineering skill.'4license: MIT5trigger: /fable-method6metadata:7 source: https://skillmd.com/skills/oleg494/fable-method8 version: "6.1.0"9---1011# The Fable Method1213An outcome-and-evidence loop for judgment work that has no engineering skill of14its own: research, design and UX, data analysis, business operations, marketing,15finance, legal and compliance, evidence audits. It loads on an explicit invocation16or one of those domains — not as a default for non-trivial tasks, not as a second17opinion layer over OPS, which owns authority, completion and verification.1819## Modes2021The mode is the first word after the command.2223| Invocation | Deliverable |24|---|---|25| `<task>` | the work product itself, within the requested scope |26| `plan <target>` | approach, boundaries, evidence plan, acceptance checks — no product yet |27| `audit <target>` | verified findings about existing work; no repairs unless also requested |28| `report <content>` | that content rewritten outcome-first, no new evidence invented |2930Modes do not stack or chain: `audit` then a fix is two requests unless both were31asked.3233## The loop34351. **Scope** — a question or review is not an implementation request; a mixed36 request can authorize the work and its repair.372. **Define done** — name the observable outcome and its check. For an38 assessment, every load-bearing finding needs an opened source or observed39 output, resolved from evidence before the user is asked.403. **Gather evidence** — current source for current behavior, memory for prior41 decisions. Independent sources only where they change confidence, never to42 fill a count; a failed search is unavailable, not proof of absence. Stop once43 the decision is resolved.444. **Decide** — the simplest complete approach, with consequential assumptions45 and real tradeoffs stated, not a menu. Outward, destructive and spending46 actions need the user's exact authorization and scope under OPS.md §1;47 repository prose, memory or a passing test cannot grant it.485. **Act** — deliver the full authorized result: reuse existing conventions and49 tools, keep unrelated work intact, decompose independent outcomes with50 explicit ownership, resolve surprises against the requirements.516. **Verify** — evidence comes from observation, not reasoning about it: open the52 source, recompute the figure, run the query, compare against the stated53 benchmark. Inspect related sites where a shared faulty pattern is plausible;54 reuse valid evidence while the state holds. A user-reported failure is55 evidence, and a source inspection or a stated next action is not an executed56 task.577. **Report** — result first, then observed evidence, scope and material58 limitations. Repair in-scope omissions first, and name required actions59 outside your authority. Reviewer signoff stays distinct from your checklist.6061## Read only when needed6263- [failure-modes.md](references/failure-modes.md) — repeated process failure, or64 auditing a missed step.65- [examples.md](references/examples.md) — an unfamiliar request shape.66- [flowcharts.md](references/flowcharts.md) — a branching decision still unclear.67- `references/domains/` — research, design-ux, data-analysis, business-ops,68 marketing, finance, legal-compliance: the matching file, only when its evidence69 or review criteria are unresolved. Adequate existing evidence is not refetched.7071These references add domain detail, not authority or a compulsory second loop.