… reference rule

Adversarial review findings (C1, C2, C3, C5, A8, F3):
- "never restate" did not cover paraphrase/summary — the actual failure
  mode in the RED evidence; now "never restate, paraphrase, or summarize".
- The No Placeholders intra-plan repetition mandate gave a symmetric
  argument for re-inlining the spec; the rule now draws the line:
  repetition WITHIN the plan is required, copying FROM the spec is not.
- Drift argument was invertible ("snapshot to avoid drift"); now states
  snapshots hide drift.
- **Spec:** header gets a no-spec branch (state requirements once in
  the header, not per task) instead of inviting "no spec, rule is moot".
- Brainstorming path bullet: an existing differently-named docs dir is
  not a "user preference" override.
- Execution Handoff now notes review fanout scales (forward-ref to
  SDD's Proportionality rule) instead of promising unconditional
  two-stage review.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Eval-caught regression: the no-spec branch added to the **Spec:**
header gave the agent a sanctioned path to skip the spec doc entirely
("avoiding duplication by skipping the spec" —
cost-spec-plan-duplication-claude-20260610T213934Z-8e5b, fail). The
branch is now scoped: if brainstorming happened the spec exists and
must be cited; "none — requirements:" applies only when requirements
arrived conversationally and no spec doc was ever produced. The
reference-discipline paragraph states the same rule up front.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

subagent-driven-development mandated implementer + two-stage review +
final reviewer unconditionally — agy and opencode each dispatched 4
subagents for a one-line console.log in the 2026-06-09 quorum sweep
(cost-trivial-task-review-fanout), and the agents that passed did so
only by disobeying the skill.

- Proportionality rule: when the entire plan is one trivial,
  fully-specified mechanical change, implement directly, verify,
  commit — no review fanout. "When in doubt, it is not trivial."
  Within a multi-task plan the full pipeline still applies to every
  task regardless of size.
- Flowchart gets the trivial-exit diamond (the failing agents follow
  the flowchart literally; prose alone would not redirect them).
- Red Flags "never skip reviews" amended to reference the exception so
  the skill does not contradict itself.

TDD evidence (quorum):
- RED: agy 025324Z + opencode batches — 4 dispatches for 1 line
- GREEN: cost-trivial-task-review-fanout-opencode-20260610T002518Z-f3f5
  pass — 0 dispatches, $0.04, change landed on main checkout
- Canary: sdd-rejects-extra-features-claude-20260610T002901Z-458a pass —
  multi-task plan still runs implementer + two-stage review per task
  (tool-called Agent ✓, spec reviewer as YAGNI gate after each task)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…es cited specs

Adversarial + consistency review findings (B1, B2, B3, B5, F1):
- Red Flags line read literally licensed skipping reviews on trivial
  tasks INSIDE multi-task plans; now states the only exception is a
  whole-plan trivial change and never-skip within multi-task plans.
- "a one-line edit" example blessed one-line behavioral changes
  (e.g. adding "|| user.isOwner"); dropped. Trivial is now defined as
  a property of the diff (no logic/control-flow/behavior change), not
  of the plan's self-description. The "nothing for review to catch"
  justification proved too much; replaced with the cost argument.
- "verify it" was undefined on the trivial path; now concrete (run
  tests/command, confirm output, verification-before-completion).
- Flowchart diamond now matches the prose: "fully-specified" + "any
  doubt = no" (the failing agents execute the flowchart literally).
- New Spec Context section + prompt-template updates: the controller
  reads the spec cited in the plan header and pastes cited sections
  into implementer/spec-reviewer prompts; the spec reviewer's
  diff-only rule gets a spec-document exception. Without this, the
  stack's reference-not-restate rule starves the SDD pipeline of
  requirements.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…ualify constant bumps

Staff-review findings (4-reviewer panel):
- CONTRADICTION FIX: Spec Context said "Subagents never read the spec
  file themselves" while spec-reviewer-prompt grants exactly that
  access. Now: implementers never read it; the spec reviewer may, at
  the cited path.
- "a constant bump" was an unqualified trivial example — a one-line
  BCRYPT_ROUNDS or session-TTL change is a security-posture change;
  now qualified "with no security or behavioral consequences"
  (matching brainstorming's config-change qualifier). The diff-property
  definition adds "nothing security-relevant".
- Proportionality rewritten 146→~115 words (house style; one statement
  of the multi-task containment instead of two).
- Red Flags Never-line trimmed 33→14 words (pointer to Proportionality
  instead of third in-file restatement).
- Prompt-template rationale tails cut (the controller just read Spec
  Context; subagents need the pasted text, not the policy rationale).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…gn (SUP-333 #3)

The HARD-GATE ("EVERY project regardless of perceived simplicity") plus
the anti-pattern list naming "a config change" made design+approval
mandatory even for fully-specified trivial asks — all 6 agents in the
2026-06-09 quorum sweep ran a multi-option design flow for "a basic
checkbox, nothing fancy" (cost-checkbox-over-trigger failed 6/6).

Two layers, because routing happens before skill content is read
(GREEN attempt 1 proved it: the agent invoked the skill on the
description's mandate and only then saw the in-skill exception, and
the invocation itself is the cost event):

- description: carve-out visible at skill-selection time — zero open
  design decisions, fully specified trivial change → implement
  directly without invoking.
- HARD-GATE: matching exception with objective re-gating tripwires
  (new file/dependency, schema/API/data question, >1 plausible
  interpretation, user frames it as a feature/project), and the
  anti-pattern section now distinguishes "seems simple" (a
  rationalization when decisions exist) from "contains every decision"
  (the exception). "A config change" moves from the all-of-them list
  to the exception's example.

The repo's acceptance test ("Let's make a react todo list" must
auto-trigger brainstorming) is unaffected: a react todo list leaves
many decisions open and todo lists remain in the anti-pattern list.

TDD evidence (quorum):
- RED: cost-checkbox-over-trigger fails 6/6 agents (batch 2026-06-09);
  GREEN attempt 1 with in-skill exception only: still fail (invoked
  via description, then asked a clarifying question)
- GREEN: cost-checkbox-over-trigger-claude-20260610T004320Z-a30e pass —
  no brainstorming invocation, agent cited the exception verbatim,
  checkbox landed in 31s
- Canary: cost-spec-plan-duplication-claude-20260610T004506Z-22ea pass —
  a real feature still triggers the full brainstorm→spec→plan flow
  (and the stacked writing-plans reference discipline holds)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…+ rationalization counters

Adversarial review findings (A1-A7, D3):
- BLOCKER A1: the re-gating tripwires lived only in the HARD-GATE, but
  the skip decision happens at the description (our own GREEN-attempt-1
  evidence). The description now carries the tripwires: adds a
  file/dependency, touches schema/API/persisted data, deletes or
  disables anything, alters behavior/security posture, >1 plausible
  reading.
- A2: "a schema/API/data question" was defeated by "the user answered
  the question"; now touch-based ("even if the user stated the desired
  outcome").
- A3: destructive changes and behavior/security-visible changes had no
  tripwire (pure removals were structurally invisible); both added.
  "a literal config value change" example now qualified ("with no
  security or behavioral consequences").
- A4: the checkbox example no longer teaches hedge-phrase = fully
  specified ("where the context leaves nothing to choose").
- A5: "EVERY project regardless of perceived simplicity" now ends
  "with exactly one exception below" instead of contradicting it.
- A6: rationalization table added (codebase-pattern, infer-the-obvious,
  hedge-phrase, asking-wastes-time).
- A7: anti-pattern opener is a claim again ("Anything with open
  decisions goes through this process").
- D3: exception states TDD and verification-before-completion still
  apply, so the fast path does not read as zero-oversight.

Description: 689 chars (limit 1024), YAML-validated.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…tion

Staff-review findings (4-reviewer panel):
- The tripwire list existed twice in this file (description +
  HARD-GATE) and the copies had already drifted after one editing
  round — the framing tripwire and the security qualifier lived only
  in the HARD-GATE, which the skip decision never reads (our own
  GREEN-attempt-1 evidence). The description is now the single
  authoritative list; the HARD-GATE exception defers to it.
- Security-posture fix: the "beyond the literally stated value" escape
  no longer applies to security — touching auth, sessions, permissions,
  CORS, or crypto re-gates EVEN when the value is exactly as stated
  (the harm of "set CORS to *" IS the stated value). User-visible
  behavior keeps the beyond-the-stated-change scope (a requested
  checkbox is the stated change; that is the point of the exception).
- The framing tripwire moves into the description where it can act.
- Anti-pattern final clause cut (was the 4th in-file statement of the
  exception's condition).
- Description: 886/1024 chars, YAML-validated.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Eval-caught leak (cost-remove-export-boundary-claude, first run): the
agent reasoned "the user already decided the deletion, so no design
decision is open" and silently removed a working feature — reading the
tripwires as indicators of open decisions rather than unconditional
re-gates. The deletion tripwire now carries the same rider as the
security one ("even when the deletion is exactly what was asked"), and
the rationalization table counters the exact quoted escape.
Description: 950/1024 chars.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…uting rule (SUP-333 #4)

Adversarial review findings D1/D2: the 1%-chance invocation rule and
the "Add X doesn't mean skip workflows" line contradicted the new
brainstorming description exception in both directions — a compliant
agent re-imposes the cost failure (invocation itself is the measured
cost event), while a cost-optimizing agent could treat any skip as
sanctioned. The routing skill now states: a documented exception in a
skill's own description defines that skill's scope (compliance, not
rationalization); any doubt about the exception's conditions means
invoke; and only the description can define one — agents cannot infer
exceptions.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…ception rule; writing-skills carve-out

Staff-review findings (4-reviewer panel):
- The skill_flow digraph still routed "yes, even 1%" straight to
  invoke with no exception branch — and this stack's own evidence says
  agents follow flowcharts literally. The flow now passes through
  "Skill's own description exempts this request?" with no/any-doubt →
  invoke.
- The <EXTREMELY-IMPORTANT> block ("you cannot rationalize your way
  out of this") read unconditional; one parenthetical defers to The
  Rule's single carve-out without weakening the block.
- Trimmed the redundant "the description defines the skill's scope"
  clause from The Rule paragraph.
- writing-skills' "descriptions must not carry process" doctrine would
  have had a future editor strip the brainstorming exception and
  silently regress the cost evals; it now distinguishes negative
  triggering conditions (scope — allowed and, per the routing rule,
  required at the description) from workflow summaries (still
  forbidden).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Measured boundary leaks (session-timeout 1/3, remove-export 1/2 on
claude) traced to silent skips: the agent acted without consulting the
exception, and nothing made the consultation observable. The skip now
requires a stated one-line scan before the first action ("Skipping
brainstorming per its exception: no security/deletion/schema/new-file
tripwires; outcome fully specified") — externalizing the scan forces
it to happen. Flowchart routes the exempt path through the scan
statement; red-flags table counters "too trivial to scan".

The trivial path stays fast: the ceremony is one sentence, not a
design flow.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>