phax-decide-spec
Skill2 starsChanged 5 months agoApache-2.0View on GitHub
Default arbitration doctrine for `phax artifact decide` on a spec — how to answer a spec's §9 open questions when no human is in the session; adopt the recommendation unless a named principle says otherwise, argue for the abandoned option first, escalate what you may not decide, and emit decisions as data.
In this repository16 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- boundariesSkill
- cli-view-layerSkill
- effect-servicesSkill
- find-docsSkill
- infrastructure-adaptersSkill
- model-routingSkill
- observabilitySkill
- phax-cliSkill
- phax-decide-planSkill
- phax-decide-reviewSkill
- phax-planningSkill
- phax-reviewSkill
- phax-specSkill
- state-machinesSkill
- validation-boundariesSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 39
Every id below ranks under one of the author's five pairs, and all five stand on one ground: E0 explicit over implicit — attribution needs …
- Distinctive: line 51
Before 1.0 there is no stability contract: break freely so the surface stays *coherent*.
- Very specific: line 67
Ask who consumes: what a human reads falls under C5 (minimal, for a named reader).
- Very specific: line 164
Every field is required; an empty why is invalid, and a verdict that departs from the recommendation cites the principle it rests on.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Bundled default doctrine, landed by hand on 2026-09-24; wired into the skill catalog by the
artifact-decideplan. A project extends it with--doctrine <file>or replaces it with a project-scoped skill of the same name.
phax decide — spec doctrine
You are arbitrating the open questions of a spec (§9). Each question is
{ id, question, options: [{ id, label, abandons }], recommendation, rationale }: every
option names what it abandons, and recommendation is the id of one option. The spec
keeps two sections, open (openQuestions) and arbitrated (arbitratedQuestions);
a decided question moves from the first to the second with its verdict. You did not write
the spec; do not defend it. Verdicts are data (see Output); phax validates them and
renders the Markdown. The framing questions F1–F5 come from decide's prompt, not from
this doctrine.
How to decide, in order
- Read the loss, not the pros. Each option names what it abandons. If an option's loss is not named, name it yourself before weighing anything.
- Argue for the abandoned option first (
S0). Write one sentence for the option the recommendation gives up, as its strongest advocate would. Only then decide. - Adopt the recommendation (
S1) unless one of the principles below says otherwise. The history behind this doctrine shows most questions resolve to their default; a deviation needs a cited principle, never taste. - Deviate only toward the stricter or more complete option (
S2). A deviation extends a principle or makes it total; it never creates an exception or a convenience. - Escalate what is not yours (
S9): a question that changes the artifact's public surface (commands, flags, config keys, file formats, APIs, UI) beyond what the spec already claims, or that reverses an earlier decision, is escalated — adopt the recommended default provisionally, markescalate, and say what a human must confirm.
The ground and the five principles
Every id below ranks under one of the author's five pairs, and all five stand on one
ground: E0 explicit over implicit — attribution needs explicit things; decisions are
dated, what is abandoned is written, "deliberately not" is recorded.
C1consistent over new — the existing default, discipline, source or gesture before a new one.C2strict over loose — refuse rather than warn; deviate only toward the stricter.C3exhaustive over convenient — total on inputs and states; write what is left out.C4condition over control — a condition the system verifies rather than a procedure a human drives; structural impossibility rather than a check.C5surface convenience over core convenience — the reader's and consumer's ease before the implementer's; never an arbitration "because it is simpler in the code". Before 1.0 there is no stability contract: break freely so the surface stays coherent; from 1.0 the contract is signed and the surface stays stable (semver). At every boundary the surface states the need and the core may propose what costs the surface nothing (phax-planning, Boundary contracts).
Attention is the cost of reading, not writing. Writing is free and there is no team
memory; a human reads code they did not write. Strict, exhaustive, explicit form does not
spend that attention, it frees it: every attributed guarantee is something the reader
need not verify. So the five pairs need no "limit on rigor"; the only line to draw is
C5's — exhaustive where a machine consumes, minimal where a human reads (surfaces,
handoffs, docs, the PR's tables).
Precedence when two pairs conflict.
E0is eligibility, not a tie-breaker: an option that leaves something implicit is not a candidate.- Ask who consumes: what a human reads falls under
C5(minimal, for a named reader); what a machine consumes falls underC3(exhaustive). They partition, they do not conflict. C2beatsC1only when the existing thing is loose — a principle is violated; otherwiseC1, the existing default stands. A refusal that widens the artifact is not "strict", it belongs to another artifact.- Among what remains,
C4: the option expressed as a condition the system verifies over one that needs control.
Principles (cite by id)
S1(C1) Adopt the default. The recommendation stands unless a principle is violated.S2(C2) Deviate toward completeness. Stricter, more total, more explicit — never more convenient.S3(C2) Refuse, don't warn. An invalid or inert input is refused with an actionable message. A warning is a decision not taken. (Refusals that widen the spec belong to the spec that motivates them, not this one.)S4(C1) One source, one gesture. Two mechanisms for the same thing is one too many; keep the one with a single source of truth. Two consents for the same loss are one consent.S5(C3) Total on inputs, minimal on outputs. A state or input that can exist is covered even if nobody produces it; an output field or flag is added only for a named reader.S6(C2) No scalar that erases attribution. A score, tier or percentage that collapses an axis the reader needs is refused.S7(C2) Nothing destructive without a separate, consented gesture. Move and keep by default; delete is its own command.S8(C1) Dissolve contradictions. When two requirements contradict, ask why the contradiction exists before choosing a side; the answer usually redraws the feature.S9(C4) Escalate surface and reversals. See step 5.S10(C5) Words for the reader. Human-facing names and messages are chosen for the person who reads them; internal names stay in--json.
Not choosing: propose and escalate, in three modes
Choosing is not the only move. When no option fits — the question is badly posed, a finding is wrong, an option outside the menu is better — the arbiter proposes and escalates: it writes its proposal (another option, a better question, a dismissed finding with its reason) and asks for a human arbitration. Its proposal never wins by itself. At equal value, prefer the option cheaper to reverse (reversal is a cost that grows toward production; only the destructive is a floor). Three modes:
- interactive — the human decides everything, the agent facilitates. The doctrine is written for this mode first: arbitration is where human attention belongs.
- headless — the agent decides within the menu and escalates what is off-menu, what changes the public surface, what reverses an earlier decision.
- verdicts —
phax artifact decide --verdicts: not the agent-decidesheadlessmode above. A person has already chosen, and the writer transcribes their choices. For each verdict it writes the advocate, the principles, the reversibility and, for a departure, the body revisions the choice implies andrevised. A verdict that chooses the recommendation after a departing verdict restores it: the writer undoes that departure's body revisions and names the sections it restores inrevised([]when the departure revised nothing; either when the departure records no revised sections, as a hand-authored Decision block does). Any other verdict that follows hasrevised: []. phax setswhy(the person's note, or the question's rationale when there is none),byandat. The writer never adds an option and never changes a choice. - headless total — later, not implemented: no escalation; the agent decides; its reservations and proposals are recorded as information for the reader; the one floor is that a destructive decision stops instead of being taken. What the headless mode escalates in practice will say whether this mode is wanted and with which floor.
In every mode the proposal is recorded (escalate or note), so a mode without
escalation defers the best move instead of losing it. Escalations are counted by the
caller; an arbiter that escalates everything is visible.
Output
A decided question keeps its id, question, options, recommendation and
rationale, moves to arbitratedQuestions, and carries a decided verdict:
{ "id": "Q1", "question": "…", "options": [ … ], "recommendation": "keep",
"rationale": "…",
"verdict": { "kind": "decided", "chosen": "keep", "abandoned": ["drop"],
"why": "…", "advocate": "The strongest case for drop: …",
"principles": ["S1", "S5"],
"reversibility": "cheap" | "costly" | "irreversible",
"revised": [] },
"history": [] }- The choice:
chosenis an optionid, or the reservedother, implicit on every question and never listed among the options;abandonedis exactly the other listed option ids, so every listed option whenchosenisother. - The reasons:
whysays why;advocateis the strongest case for an abandoned option (S0). - The citations:
principleslists ids that open list items in the loaded doctrine (this skill and any--doctrinefile); an id named only in prose, likeS0, is not citable. - The cost of going back:
reversibilityischeap,costlyorirreversible. - The body:
revisednames the body sections the verdict changed, empty when none. A verdict whosechosendeparts from the recommendation revises the body it implies (requirements, surface, acceptance criteria) and names it here. - The attribution: phax sets
by(the operator, by the committing git identity) andat; a value given there is overwritten. - The record: a re-arbitrated question keeps its previous verdict at the end of
history.
Every field is required; an empty why is invalid, and a verdict that departs from the
recommendation cites the principle it rests on. Never edit the spec's requirements to make a question disappear. The headless
modes above record a proposal as escalate or note; the interactive verdict has no such
field — the operator decides.
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.

