foundation-stakeholder-briefings
Turns any source artifact (spec, discovery, research, GTM plan, experiment results, retro, or raw notes) into one canonical master document plus a set of audience-tailored briefings, each re-pitched to a stakeholder lens (executive, board, engineering, UX, PMM, sales, CS, legal, data, or a custom audience). Every briefing is a traceable projection of the master, so the versions never disagree. Use when one piece of work must reach several audiences who each need a different framing, decision, and level of detail.
git clone --depth 1 https://github.com/product-on-purpose/pm-skills /tmp/foundation-stakeholder-briefings && cp -r /tmp/foundation-stakeholder-briefings/skills/foundation-stakeholder-briefings ~/.claude/skills/foundation-stakeholder-briefingsSKILL.md
<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 --> # Stakeholder Briefings A stakeholder-briefings artifact takes one source (a PRD, a discovery synthesis, a research report, a GTM or launch plan, experiment results, a retro or incident write-up, or raw notes) and produces a single saveable file containing: 1. a **master document**: the canonical, audience-neutral synthesis of the work (what and why, decisions, status, risks and open questions, asks, timeline), with each claim numbered (`M1`, `M2`, ...); and 2. a set of **audience briefings**: the same content re-pitched for each chosen stakeholder, one self-contained, copy-paste-ready block per audience. The skill runs **master-first, then projects**. The master is the single source of truth; every briefing is a projection of it. A briefing may omit, reorder, and translate master content, but it may never assert a claim that is not in the master. That projection rule is what keeps the executive version and the engineering version from quietly disagreeing, and it is the difference between this skill and asking a model to "rewrite this six ways." Distinct from `foundation-stakeholder-update` (one async update of meeting outcomes for a single audience, meeting-bound), `discover-stakeholder-summary` (a map of who stakeholders are and their influence/interest), and `foundation-persona` (a customer/buyer viewpoint to design or market against). ## When to Use - One piece of work must reach several audiences who each need a different framing, decision, and level of detail (a spec going to engineering, design, data, and the funder at once). - You are about to manually rewrite the same update three to five ways, one per audience. - A decision or result needs to propagate across functions without the versions drifting apart. - A single audience needs a tailored briefing from a non-meeting source (N=1 is supported; the fan-out is the signature use, not a floor). ## When NOT to Use - One async update of **meeting** outcomes for stakeholders. Use `foundation-stakeholder-update` (it is meeting-bound; that is its scope). - Understanding or mapping stakeholders (influence, interest, comms plan). Use `discover-stakeholder-summary`. - A persona to design or market against. Use `foundation-persona`. - There is no source content yet. This skill projects an existing artifact; it does not do the underlying analysis. ## Instructions When asked to create stakeholder briefings, follow these steps: 1. **Ingest and classify the source.** Read the provided artifact. Classify its type (spec/PRD, discovery/research, GTM/launch, strategy/roadmap, experiment/metrics, incident/retro, compliance/privacy/security, or raw/ambiguous). If the source is thin, continue but set `input_quality: low` and name the gap. 2. **Build the master.** Write the audience-neutral canonical document with these sections: What and Why, Decisions, Status, Risks and Open Questions, Asks, Timeline. **Number every load-bearing claim** with a stable ID (`M1`, `M2`, ...). The master carries no audience-specific spin; it is the shared substrate. 3. **Propose the audiences.** From the source type, propose the relevant subset using `references/source-type-map.md` (for example, a spec proposes Engineering, UX/Design, Data/BI, Executive; a GTM plan proposes PMM, Sales, CS/Support, Executive). Present the proposal and accept `go` (generate the proposed set), an edit (`drop X, add Y`), or `all` (all nine). If invoked with `--go`, skip the prompt and generate the proposal. No audience is ever locked out. 4. **Project each briefing.** For each chosen lens (see `references/audience-lenses.md`), render a self-contained block delimited by `--- BEGIN: <lens> ---` / `--- END ---`, containing: - `Draws on:` the master claim IDs this briefing projects (required). - `Primary ask:` exactly one decision or action for this audience (required). - a one-line headline, a "what this means for you" framing, and the body, at the lens's length, vocabulary, and tone. Every load-bearing line must trace to a master claim. Do not introduce a claim that is not in the master. 5. **Flag translations.** Keep a translations-applied log (internal, below the shareable boundary) for every technical-to-business or inferred re-pitch, so the user can verify it lands. This section is never part of a shareable briefing. 6. **Self-check the invariant** before finalizing: - **Trace references resolve** (deterministic, checkable): every briefing `Draws on:` ID resolves to a real master claim. - **One CTA** (deterministic, checkable): exactly one `Primary ask:` per block. - **No untraced claim** (review): re-read each block against its `Draws on:` set and confirm the body introduces nothing absent from those master claims. This is a review step, not automated. - **Neutral master** (review): the master has no audience-specific spin. List anything that fails. 7. **Render the artifact.** Master (with claim IDs) -> the delimited briefing blocks -> the boundary marker -> the translations-applied log -> Sources and References. Remove all guidance blockquotes from the final output. ## Audience lenses Nine first-class lenses, each defined by the decision it owns, plus a Custom slot whose lens is inferred from the audience name and source and shown for confirmation. Full definitions, per-lens "not this lens when" boundaries, and the overlap matrix (Exec vs Board, PMM vs Sales, Engineering vs Data, Legal vs Exec) are in `references/audience-lenses.md`. ## Output Format - A single artifact (filename `YYYY-MM-DD_HH-MMtz_<title>_stakeholder-briefings.md`), built from `references/TEMPLATE.md`. - Each briefing block is self-contained and send-ready (BEGIN/END cut-lines) so it can be copied out without edits. - A future `--split` mode (write each block to its own file) is deferred; v1 is single-artifact. ## Quality Checklist - [ ] Master present with numbered claim IDs (`M1`, `M2`, ...
Run an ad-hoc ordered chain of pm-skills with shared context (ephemeral; routes to the pm-workflow-orchestrator)
Defines a testable hypothesis with clear success metrics and a validation approach. Use when forming assumptions to test or aligning a team on what success looks like, before any experiment is designed. To design the A/B test or experiment that will validate the hypothesis, use measure-experiment-design.
Creates a Jobs to be Done canvas capturing the functional, emotional, and social dimensions of a customer job. Use when deeply understanding customer motivations, designing for jobs, or reframing product positioning.
Creates an opportunity solution tree connecting a desired outcome to customer opportunities and candidate solutions, preventing solution-first jumps in continuous discovery. Use when structuring scattered feature ideas or communicating how the roadmap traces to outcomes. For ranking an existing flat list of candidates, use define-prioritization-framework instead; this skill discovers the list, it does not score one.
Run applicable prioritization frameworks (RICE, ICE, MoSCoW, Weighted Scoring, Kano) against a list of features or initiatives. Produces a comparison table showing where rankings agree and diverge across frameworks, and an executive summary with recommendation. Framework applicability is filtered by data availability; Kano requires customer research. Refuses to fabricate scores; produces an estimation scaffold when input data is missing.
Creates a clear problem framing document with user impact, business context, and success criteria. Use when starting a new initiative, realigning a drifted project, or communicating up to leadership.
Generates structured Given/When/Then acceptance criteria for a user story or feature slice, covering the happy path, key failure scenarios, and non-functional expectations in testable form. Use when turning requirements into verifiable scenarios for engineering handoff and QA sign-off. For a dedicated catalog of boundary conditions, error states, and recovery paths across a feature, use deliver-edge-cases; to write the stories themselves, use deliver-user-stories.
Documents edge cases, error states, boundary conditions, race conditions, and recovery paths for a feature - the systematic catalog of what can go wrong and the failure modes to design for. Use during specification to map the failure surface and ensure comprehensive coverage, or during QA planning to identify boundary and limit scenarios to test. Distinct from deliver-acceptance-criteria, which writes story-level Given/When/Then checks; this skill produces the whole-feature edge-case catalog.