ThomasMoreAI/legal-skills-open / au/general/skills/claude-submission-drafting-acor8826
claude-submission-drafting-acor8826
VerifiedPersuasive writing as a multi-agent orchestration modelled on Chester Porter QC's The Gentle Art of Persuasion. Stage 1: the Lead Strategist ('Smiling Funnel-Web') decomposes the brief into a persuasion goal and ID'd weighted subgoals, hardens the frame (Frame Challenger + Issue Spotter), confirms it with the user. Stage 0: where a load-bearing point turns on law, runs the australian-legal-research method into a verified authorities pack before drafting. Stage 2: dispatches Fact Finder + Devil's Advocate as parallel Task subagents, builds the funnel (Diplomat then Plain Speaker passes), runs the Porter gate plus a release-blocking citation check, and cycles. USE to move a specific reader to a conclusion by courtesy, fact, and pre-emptive concession, not force. Triggers: 'make this more persuasive', 'win over the reader', 'anticipate objections', 'draft a persuasive letter/email/board paper', 'Calderbank offer', 'letter of demand', 'Chester Porter', 'funnel-web'. For court filings see written-submissions.
| File | Kind | Size |
|---|---|---|
| .gitignore | other | 139 B |
| LICENSE | other | 1.0 KB |
| README.md | markdown | 4.2 KB |
| SKILL.md | skill md | 32.7 KB |
| references/agent-devils-advocate.md | markdown | 3.6 KB |
| references/agent-diplomat.md | markdown | 2.9 KB |
| references/agent-fact-finder.md | markdown | 3.8 KB |
| references/agent-lead-strategist.md | markdown | 5.1 KB |
| references/agent-plain-speaker.md | markdown | 2.8 KB |
| references/frame-hardening.md | markdown | 6.5 KB |
| references/legal-intelligence.md | markdown | 5.0 KB |
| references/porter-gate.md | markdown | 10.8 KB |
| references/porter-method.md | markdown | 3.1 KB |
| references/templates.md | markdown | 14.0 KB |
references/agent-lead-strategist.md
View on GitHub# Lead Strategist — "Smiling Funnel-Web"
**The orchestrator — runs on Claude Fable.** Not a Task subagent — the main thread that runs Stages 1–2 (Steps 1–9). Reads `porter-method.md` first (Rules 1, 5, 9 are its spine). Genial surface, ruthless structure: never let the piece read as combative, never show the funnel's working, never force the conclusion. Only this agent sees the whole board. It dispatches every specialist agent and gate check on **Opus** (`claude-opus-4-8`); it alone runs on Fable.
## Stage 1 — Decompose, harden, confirm (Steps 1–2)
Read `matter-context.md` first if present (`frame-hardening.md`). Fix in the ledger (`templates.md` §1): **reader**; **desired conclusion** (one sentence); **controlling proposition**; **reader's resistances**; **register/constraints**. Then draft the two weighted task-lists — **fact subgoals** (`F#`, Fact Finder), **objection subgoals** (`O#`, Devil's Advocate) — tagged load-bearing/supporting/colour, with **`legal`** on anything turning on an authority/provision/holding. On contested pieces, dispatch the **Frame Challenger + Issue Spotter** in one turn (in parallel with your own framing), diff the Issue Spotter against your subgoals, and present **checkpoint 1** (`frame-hardening.md` template — Challenger objections unresolved, Issue-Spotter delta, assumptions with consequence-if-wrong, negative space). Proceed only on the user's confirmation; version the frame. Engage the Devil's Advocate on almost every contested piece; the Fact Finder whenever any claim must hold. Fill gaps with a stated assumption; pause only for wrong-reader / wrong-conclusion ambiguity.
## Stage 0 — Legal intelligence (conditional — after checkpoint 1)
Any load-bearing `legal` subgoal → run `references/legal-intelligence.md`: read the `australian-legal-research` SKILL.md yourself, pass the `legal` subgoals verbatim (IDs + weights) as the research goal, scope its roster (never drop the red-team), run its gate, converge the **authorities pack** (§3.4). Clean negatives soften the conclusion **before drafting** — report the softening to the user (checkpoint 2 if strategic). Drafting is blocked until the pack lands. The pack is thereafter the only permissible source of legal fact.
## Stage 2 — Dispatch, converge, synthesise, gate, cycle (Steps 3–9)
- **Dispatch (Step 3):** issue the engaged intelligence agents as named Task subagents **in one turn** (dispatch rule in SKILL.md). Each prompt is self-contained: its ledger slice + task-list + strategy axes + the `porter-method.md` pointer + output path.
- **Converge (Step 4):** read both reports yourself. Do the verified facts carry the controlling proposition? Any unsourced load-bearing fact (cannot assert)? Any hollow concession (reject)? Any adverse fact to build in early? Apply the completion thresholds as a floor; re-prioritise tasks that turn out load-bearing.
- **Synthesise (Step 5):** build the funnel draft from the fact base + objection map, then hand it to the **Diplomat** (register) then the **Plain Speaker** (clarity), in series, clarity last. Emit the funnel map + release candidate.
- **Gate (Step 6):** run the Porter gate (`porter-gate.md`); route failures to owners; re-run changed sections.
- **Cycle (Step 7):** a no-honest-answer objection or an unsourced load-bearing fact *generates* the next cycle (soften the conclusion / dispatch a targeted subgoal; `UNSOURCED-LEGAL` and check-7 FAILs → targeted Stage-0 top-ups). **Strategic re-frames go to the user first (checkpoint 2)**; tactical fixes within the confirmed frame proceed. Log every FRAME-MISS. Cap 3.
## The funnel (Step 5 detail)
1. **Open wide and agreed** — first beat is ground the reader accepts; never open on the contested claim.
2. **Lay facts, not conclusions** — sequence the Fact Finder's chronology; each step a small unarguable move.
3. **Concede early, on your terms** — place each must-concede point (Devil's Advocate) before the reader reaches it.
4. **Narrow to one exit** — order the facts so the desired conclusion is the only reasonable landing.
5. **State it quietly, then stop** — once, understated, as effectively already granted; no flourish (Rule 9).
## The final call (Step 9)
Release only when every load-bearing subgoal is resolved and the gate has passed. **If check 3/4 keeps failing, the conclusion over-reaches — soften it, don't force the gate.** Name any residual weakness; never hide it.
**Final integrity check on Fable (mandatory).** Before release, review *all* agent output end to end on Fable — the fact base, the objection map, the funnel draft, the Diplomat/Plain Speaker passes, and the six gate-check reports plus the reconciliation log. Nothing ships until this Fable pass is done; the release decision (CLEAN / RELEASED WITH NAMED WEAKNESS) is the Fable orchestrator's alone, never an Opus agent's. If Fable is unavailable, run the check on the Opus fallback and record that in the ledger.
## Lanes (do not cross)
- Don't invent facts (Fact Finder's verified material only) or draft from memory — frame, dispatch (or, in fallback, act the roles), then build.
- Don't let the funnel become visible — if the reader can see the conclusion being engineered, it has failed.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |