All skills
simota avatar

/scribe

@35ffd55
by shingo imotasimota/agent-skills85 stars
15

Authoring standalone and cross-team specifications: PRD/SRS/HLD/LLD, staged L0-L4 unified packages, BDD acceptance criteria, and traceability. Use for technical or multi-audience documentation; not implementation or architecture decisions.

Use this Skill: https://skilld.dev/gh/simota/agent-skills/scribe

This session only. Nothing lands on disk.

referenceunified-specstakeholder-map.md

≈2k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Stakeholder Map Reference

Purpose: Map stakeholders on a Power × Interest grid, assign an engagement mode per quadrant, and design communication cadence and information-flow patterns. Scribe uses this to anchor the L0 Vision audience and to drive communication plans for L1-L3 review cycles.

Scope Boundary

  • scribe stakeholder: Power/Interest grid + engagement design (this document).
  • scribe vision (elsewhere): L0 Vision Block — stakeholder map informs the audience section.
  • scribe raci (elsewhere): Per-item responsibility assignment (different question).
  • cast (elsewhere): User personas. Stakeholder ≠ user; an admin, a regulator, a vendor, a board member are stakeholders but not users.
  • saga (elsewhere): Narrative for external audiences.
  • Magi (elsewhere): Multi-perspective arbitration.

Mendelow Power × Interest Grid

          HIGH INTEREST
               │
    Keep       │    Manage
    Informed   │    Closely
               │   (Key players)
               │
───────────────┼───────────────── HIGH POWER
               │
    Monitor    │    Keep
    (min effort)│    Satisfied
               │
          LOW INTEREST

Four quadrants

Quadrant Characteristics Engagement mode
Manage Closely (high power, high interest) Decision-makers who care Deep involvement, co-creation, weekly touchpoints
Keep Satisfied (high power, low interest) Executives, compliance, legal — rubber-stamp when happy Concise updates, headline-level only, avoid surprises
Keep Informed (low power, high interest) Users, champions, power users Regular updates, include them in feedback loops
Monitor (low power, low interest) Tangential teams, occasional users Minimal communication, notify on launch

Identification Workflow

LIST         →  enumerate every party that influences or is influenced
             →  internal: teams, functions, execs, board
             →  external: customers, vendors, regulators, partners
             →  silent: future users, legal/ethics, underrepresented

SCORE        →  power: can they block / accelerate / shape outcome? (1-5)
             →  interest: how directly does it affect their work? (1-5)

POSITION     →  place on 2x2 grid by median score cutoff
             →  watch for quadrant ties → tiebreak by recent activity

ENGAGE       →  assign engagement mode per stakeholder
             →  define information artifact (memo, demo, dashboard)
             →  define cadence (weekly / biweekly / monthly / at-milestone)

LOOP         →  reassess each quarter (positions change)
             →  new stakeholders enter; power shifts

Engagement Mode Matrix

Quadrant Artifact Cadence Owner
Manage Closely Working sessions, draft specs, demos Weekly PM + Lead
Keep Satisfied Executive summary, dashboard, status email Biweekly / monthly PM
Keep Informed Release notes, community posts, office hours At milestone DevRel / Marketing
Monitor FYI channel post, release note link At launch Auto

Influence Dynamics

Not every stakeholder on the grid operates independently. Model:

  • Coalitions: groups that move together (e.g., customer advisory board).
  • Blockers: single stakeholders with veto (legal, security review).
  • Champions: stakeholders who advocate internally on your behalf.
  • Silent stakeholders: absent from the room but affected (end users, future hires, regulators).
  • Proxy stakeholders: represent others (customer success rep represents users).

RACI-adjacent roles (optional dimension)

For decision-heavy contexts, annotate per stakeholder:

Role Meaning
D Driver (moves work forward)
A Approver (signs off)
C Contributor (provides input)
I Informed (kept in loop)

This dovetails with scribe raci.

Information Flow Patterns

Different stakeholders need different views of the same reality.

Pattern Example
Top-down summary Exec sees 1-pager monthly; team sees full doc
Bottom-up signal Users report via NPS; PM aggregates for leadership
Peer-to-peer bridge Design shares Figma with engineering daily
Gate-and-review Legal reviews only at milestone boundaries
Dashboard pull Stakeholder self-serves from live dashboard
Milestone push Launch + post-mortem sent to all at event

For each high-power stakeholder, choose one dominant pattern. Mixing causes noise.

Output Template

## Stakeholder Map: [Project / Feature]

### Stakeholder Inventory
| Stakeholder | Power (1-5) | Interest (1-5) | Quadrant | Role (D/A/C/I) |
|-------------|-------------|----------------|----------|----------------|
| [name/role] | [n] | [n] | [quadrant] | [role] |
| ... | ... | ... | ... | ... |

### Grid Visualization
[ASCII or Canvas handoff]

### Engagement Plan
| Quadrant | Stakeholders | Artifact | Cadence | Owner |
|----------|-------------|----------|---------|-------|
| Manage Closely | [list] | [doc type] | [freq] | [person] |
| Keep Satisfied | [list] | [doc type] | [freq] | [person] |
| Keep Informed | [list] | [doc type] | [freq] | [person] |
| Monitor | [list] | [doc type] | [freq] | [person] |

### Dynamics
- **Coalitions**: [...]
- **Blockers**: [...]
- **Champions**: [...]
- **Silent**: [...] — how surfaced in decisions

### Information Flow
| Stakeholder | Dominant pattern | Why |
|-------------|------------------|-----|
| [name] | [pattern] | [reason] |

### Risks
- [ ] Blocker identified with mitigation plan
- [ ] Silent stakeholders have a surrogate or representative
- [ ] Executive Keep-Satisfied group has pre-agreed surprise policy
- [ ] Cadence feasible with team bandwidth

### Review
- **Next reassessment**: [quarter / trigger event]

### Handoffs
- Scribe L0 Vision: audience section references this map
- Scribe `raci`: per-item responsibility assignment
- Prose: communication templates per artifact
- Canvas: visual rendering of the grid
- Scribe: formal stakeholder register document

Anti-Patterns

Anti-pattern Fix
Everyone listed as "manage closely" Apply discipline; the grid has four quadrants for a reason
Static map, never updated Reassess quarterly; after org change or pivot
Conflating stakeholders with users Users get research/persona; stakeholders get engagement plans
No silent-stakeholder analysis Add proxy representation; future users, regulators, ethics
One-size-fits-all cadence Power/Interest quadrant drives cadence
No named owner per engagement Assign explicitly; "the team" = no one
Ignoring blockers until milestone Surface blockers in discovery; give them early warning
No conflict-resolution plan When stakeholders disagree, pre-define escalation

Deliverable Contract

When stakeholder completes, emit:

  • Stakeholder inventory with power × interest scores.
  • Grid visualization (ASCII or Canvas handoff).
  • Engagement plan per quadrant (artifact, cadence, owner).
  • Dynamics (coalitions, blockers, champions, silent).
  • Information flow pattern per high-power stakeholder.
  • Risk check and review cadence.
  • Handoffs: Scribe L0, Scribe raci, Prose, Canvas, Scribe.

References

  • Aubrey Mendelow — original Power/Interest matrix (1991)
  • Johnson, Scholes, Whittington — Exploring Corporate Strategy (applied matrix)
  • PMI PMBOK — stakeholder engagement processes
  • Gardner — "Effectively Managing Upwards" (Keep Satisfied dynamics)
  • Marty Cagan — Inspired (product stakeholder patterns)
  • Teresa Torres — Continuous Discovery Habits (customer as stakeholder vs user)
  • Cynefin / Stakeholder Salience Model (Mitchell, Agle, Wood 1997) — power/legitimacy/urgency extension
  • Atlassian / Miro templates for stakeholder mapping workshops

Source: SKILL.md on GitHub

1 warning13d5 checks · Risk SAFE
  • Gen Agent Trust Hub13d

    No security issues were detected. The skill is designed for authoring technical specifications and documentation. It integrates standard format conversion tools and follows industry best practices for requirements engineering and AI-agent compatibility.

  • Socket13d

    No alerts

  • Snyk13d

    Risk: LOW · No issues

  • Runlayer6mo

    1/8 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 35ffd55. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated 2 weeks ago

README badge

README badge for simota/agent-skills/scribe