菜单
Produces a ROADMAP.md (sequenced product bets with rationale + an explicit deferred set) from an attested product brief/strategy; primes the editorial + compliance gates. Use after the brief/PRD to sequence what to build when.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Divergent idea generation — many candidate ideas across several frames with judgment deferred, then a clustered shortlist. Use to open a solution space wide, or /brainstorm. Not for developing one idea (ideate) or picking one (propose).
Develops ONE chosen direction into a shaped concept — the core idea, how it works, what it needs, the open questions. Use to deepen a single idea, or /ideate. Not for generating many (brainstorm) or deciding among them (propose).
Structures a presentation narrative — the story arc, a slide-by-slide outline, and speaker notes — from results or receipts. Use to shape a talk or deck, or /present. Not for prose reports (report) or rendering visuals (asset).
Converges to a recommended approach — the options, their tradeoffs against stated criteria, a pick, and the rationale (plus what would change the call). Use to decide among approaches, or /propose. Not for generating options (brainstorm) or deep-diving one (ideate).
Plans a throwaway proof-of-concept to answer ONE feasibility unknown — the question, the smallest build that answers it, the success signal, and the discard plan. Use to de-risk before committing, or /prototype. Not for the real build (executor) or deciding (propose).
Produces a REDTEAM-FINDINGS.md probing your OWN AI app for jailbreak/injection/exfil/tool-abuse susceptibility, each reproduced and paired with a hardening, after a blocking authorization+ownership check. Use to red-team an owned AI system defensively.
| name | roadmap |
| description | Produces a ROADMAP.md (sequenced product bets with rationale + an explicit deferred set) from an attested product brief/strategy; primes the editorial + compliance gates. Use after the brief/PRD to sequence what to build when. |
The sequencing producer of the product domain — it turns an attested PRODUCT-BRIEF.md (the why/who)
and, when present, the prioritization output (the ranked what) into a ROADMAP.md (the what, in what
order, and why): sequenced bets, each with a rationale tied to the success metric, and an explicit
"deferred / not now" set. The bar is that the sequence has a stated logic (dependency / value / risk /
learning) and commitments are not overstated — horizons are confidence-tagged bets, not guaranteed
dates. Verified by Attestation: a named owner attests the sequencing is sound and owns the call; a
roadmap cannot be tested or measured at decision time. roadmap runs after strategy/prioritization, not
instead of them.
Does: resolve the roadmap sub_target, read the attested brief (and prioritization if present), and write a
ROADMAP.md of sequenced bets + per-bet rationale + dependencies + horizons + an explicit deferred set,
grounded in the brief's success metric; write a roadmap receipt. Does not: decide the problem/persona/
success metric (that is product-brief), rank the backlog under a framework (that is prioritization),
build the bets (the target domain's executor does), issue the gate verdict (the editorial/compliance gates
do — this primes them), or promise dates as guarantees (horizons are bets with confidence).
PRODUCT-BRIEF.md (and no strategy) yet — run product-brief first; a roadmap with
no problem/success contract sequences bets toward nothing.prioritization; a
roadmap sequences the chosen bets, it does not score them.ROADMAP.md still fits and nothing changed — editing it beats rewriting a commitment
stakeholders are tracking.roadmap.| Situation | Reference |
|---|---|
| The ROADMAP.md contract template | references/roadmap-template.md |
| Per-sub_target must-declares · overlay · vocabulary | ../acceptance/sub-targets.md · ../acceptance/overlay.md · ../vocabulary.md |
| The upstream product contract | the attested PRODUCT-BRIEF.md (+ prioritization output if present) |
PRODUCT-BRIEF.md (problem, persona, success metric, non-goals) — the anchoring contract.../vocabulary.md, ../acceptance/) + state/intake.json.PRODUCT-BRIEF.md; if absent, stop and route to product-brief (ordering:
strategy before the roadmap). Carry the success metric forward as the frame every bet is judged against.record_type: alert), never stated as a guarantee (the compliance gate).ROADMAP.md from references/roadmap-template.md; name the attestation owner.A gate failure is fixed before lock, not waived — these mirror the editorial/compliance gates.
product/ROADMAP.md if that dir exists) — persistence:
Spec-Anchored. Sequenced bets · per-bet rationale tied to the success metric · stated sequencing logic
· dependencies · horizons (confidence-tagged) · explicit deferred set · attestation owner.<workspace>/receipts/roadmap-receipt-<ts>.json): base schema + sub_target, bets,
sequencing_logic, horizons, deferred, roadmap_path, attested_by, gate_checks. persistence:
Spec-Anchored.product-brief first; do not sequence bets toward an unstated goal.Done when: a ROADMAP.md exists anchored to an attested brief, with sequenced bets each carrying a
rationale tied to the success metric, a stated sequencing logic, confidence-tagged horizons, an explicit
deferred set, and a named attestation owner; a roadmap receipt records the gate checks.
Not done if: there is no product brief; a bet lacks a rationale or a tie to the metric; the sequence has no stated logic; horizons are dated guarantees; the deferred set is missing; or commitments were stated as fact rather than flagged.
Next step: prioritization if the bets still need ranking under a framework; the target domain's
executor builds each bet (verified by that domain's mode); the named owner attests the roadmap (Attestation).
roadmap sequences bets
toward that attested metric. Always run product-brief first.roadmap sequences the
chosen bets into horizons; prioritization scores, roadmap sequences. Complementary.