Claude for Strategy & Transformation

Turn messy strategy and transformation work into board-credible drafts in minutes — twelve practice plugins with consulting discipline, trust guardrails, and scheduled watchers.

Install in one command

In Claude Code or Claude Cowork:

/plugin marketplace add <path-to-this-repo>
/plugin install consulting@claude-for-strategy

Restart Claude Code, then run /consulting:practice-setup. Pick user scope when asked. Full walkthrough: QUICKSTART.md.

[!IMPORTANT] Every output is a draft for your review — not a certified methodology, not a board-ready recommendation, not a substitute for your analysis and professional judgment. Source tagging on external claims, load-bearing assumptions surfaced, conservative defaults on invented inputs, and explicit gates before board- or exec-facing finals. You review, verify, and take responsibility for anything that leaves the building.

Plugins at a glance

Role-based routing (which plugin to install first): QUICKSTART.md#which-plugin-is-for-me.

Worked examples

Each example produces a draft artefact for your review — paste your context, run the command, then verify assumptions and numbers before anything goes to a board or steering committee.

1. Steering committee pack (PMO)

You have: a RAID log, milestone tracker, and two decisions pending approval.

Run: /pmo:steering-pack — paste or point at your RAID and milestone files.

You get: a decision-led steering deck outline — RAG summary, escalations, critical-path slippage, and explicit decision asks with owners. The skill reads your practice profile for governance cadence and escalation thresholds.

2. Transformation business case (Transformation)

You have: an initiative scope, rough cost/benefit notes, and no formal template yet.

Run: /transformation:business-case — describe the problem, options considered, and what you know about costs and benefits.

You get: a structured business case — problem statement, options (including do-nothing), cost/benefit with provenance tags on every figure, sequencing, risks, and a clear ask. Missing inputs are flagged INPUT NEEDED, not invented.

3. Board performance narrative (Performance)

You have: quarterly metrics in a spreadsheet or bullet list — some on track, some breaching threshold.

Run: /performance:performance-narrative — paste the metrics and name your audience (board, exec committee, functional lead).

You get: a BLUF-structured narrative — what changed, what breached, root-cause hypotheses, and recommended actions. Load-bearing assumptions and unverified benchmarks are surfaced before you stamp it final.

More agents and commands: Named agents (v2) · Extended agent catalog · Skill & command reference.

Named agents (v2)

Twelve job-titled entry points for strategy and transformation work. Each name maps to exactly one slash command under skills/<skill-name>/ — on-demand agents are not duplicated as files under agents/ (that directory is for scheduled watchers only).

Run the plugin's practice-setup before first use — every skill reads your practice profile and applies the trust spine (source tagging, surfaced assumptions, board-ready gate).

Extended agent catalog

Each agent below is named for the job it does. Start with the v2 entry points or the ones that match your work, then tune the underlying skill, practice profile, and connectors to how your team does it.

Everything here ships as Claude Cowork or Claude Code plugins and as managed-agent cookbooks for headless deployment through the Claude Managed Agents API — same skills and prompts, two surfaces from one source.

What's in the repo:

• Practice-area plugins covering consulting craft, corporate strategy, market intelligence, transformation, change management, org design, performance, PMO, balanced scorecards, OKRs, and value realisation — each built around a practice setup that learns your playbook, a living CLAUDE.md practice profile every skill reads from, and a propose profile update flow so conventions can be recorded mid-engagement without re-running setup.
• Strategy Builder Hub for discovering, installing, and QAing community strategy skills from trusted registries.
• MCP connectors across general productivity (Slack, Google Workspace, Atlassian, Linear, Asana, Monday.com) and strategy-specific categories (GitHub, Miro, observability, hosting, spreadsheets).
• Named agents (v2) — twelve job-titled entry points plus the extended catalog below.
• Scheduled agents in pmo, performance, transformation, okr, balanced-scorecard, corporate-strategy, market-intelligence, and value-realisation — escalation/slippage/steering prep, KPI breach watching, cadence reporting, assumption-decay/roadmap-drift, check-in nudges, scorecard review reminders, portfolio review prompts, competitive signal scanning, benefits-tracking checkpoints. Convention Monitor (consulting) is still planned for V1.2.

Repository layout

consulting/               # narrative, hypothesis trees, workplans, decks, memos, doc review
corporate-strategy/       # growth vectors, resource allocation, portfolio calls, strategic options
market-intelligence/      # competitive landscape, incentives, positioning, information asymmetry
transformation/           # maturity, roadmaps, TOM, tech strategy, business cases
change-management/        # stakeholder impact, readiness, sponsor roadmaps, comms, resistance
operating-model/          # structure fit, decision rights, span/layers, matrix reporting, rewards alignment
performance/              # KPI trees, trackers, glossary, performance narrative
pmo/                      # RAID, status reports, steering packs, milestones, decisions
balanced-scorecard/       # strategy maps, perspectives, measures, cascade, review
okr/                      # objectives, key results, targets, cascade, check-ins, scoring
value-realisation/        # benefits mapping, baseline discipline, tracking, recovery, PIR
strategy-builder-hub/     # community skill discovery, installation, QA, and update management
external-plugins/         # vendor plugins (future)
managed-agents/           # CMA cookbooks — escalation-watcher, steering-prep, roadmap-drift-watcher, competitive-signal-scan, realisation-checkpoint-reminder
scripts/                  # validate.py · sync-references.py · check-marketplace-sync.py · validate-skills.py · validate-submission.py · validate-connectors.py · lint-tool-scope.py · deploy-managed-agent.sh
references/               # repo-root mirror of consulting/references/ · connector-taxonomy.json
.claude-plugin/
  marketplace.json        # plugin registry
CONNECTORS.md             # contributor guide to the ~~ placeholder convention
QUICKSTART.md             # 60-second install path

Each plugin directory has the same shape:

<plugin>/
  .claude-plugin/plugin.json
  CLAUDE.md               # template practice profile — filled in by /<plugin>:practice-setup
  README.md
  skills/                 # skills — each is a /<plugin>:<skill> slash command
  agents/                 # scheduled/background watcher definitions only (if any)
  hooks/                  # pre- and post-tool hooks (if any)

Getting started

Claude Cowork

1. Open the Cowork tab.
2. Click Customize in the left sidebar.
3. Click Browse plugins and install the ones you want, or upload a custom plugin file (any plugin directory zipped up).

After install, skills fire automatically when relevant, slash commands are available via /, and scheduled agents run on the cadence set in their frontmatter.

Claude Code

# Add the marketplace (use the absolute path to this repo or a GitHub URL)
/plugin marketplace add <path-to-this-repo>

# Install a plugin — pick the ones that match your practice
/plugin install consulting@claude-for-strategy
/plugin install corporate-strategy@claude-for-strategy
/plugin install market-intelligence@claude-for-strategy
/plugin install transformation@claude-for-strategy
/plugin install change-management@claude-for-strategy
/plugin install operating-model@claude-for-strategy
/plugin install performance@claude-for-strategy
/plugin install pmo@claude-for-strategy
/plugin install balanced-scorecard@claude-for-strategy
/plugin install okr@claude-for-strategy
/plugin install value-realisation@claude-for-strategy
/plugin install strategy-builder-hub@claude-for-strategy

# Restart Claude Code, then run setup for each plugin you installed.
# This writes org-profile.md (once) and per-plugin CLAUDE.md to ~/.claude/plugins/config/claude-for-strategy/
/consulting:practice-setup
/corporate-strategy:practice-setup
/market-intelligence:practice-setup
/transformation:practice-setup
/change-management:practice-setup
/operating-model:practice-setup
/performance:practice-setup
/pmo:practice-setup
/balanced-scorecard:practice-setup
/okr:practice-setup
/value-realisation:practice-setup
/strategy-builder-hub:practice-setup

Run the practice setup first. Every other skill in a plugin reads from the practice profile it writes. Skipping setup is the single most common reason a skill produces generic output. The interview takes 10–20 minutes per plugin and will ask you to point at seed documents. More seed material is better; a quick start option is available if you want to be productive in 2 minutes and refine later.

Which plugin is for me? See the role routing table in QUICKSTART.md.

Updates: /plugin update.

How it fits together

Everything is markdown and JSON. No build step.

Claude Managed Agents

Five scheduled watchers ship as managed-agent cookbooks today — same system prompts and skills as their plugin agents, deployable headlessly behind your own workflow engine:

Roadmap X4 watch agents (Competitor-Move Monitor, Market-Signal Radar, Transformation-Programme Health Watcher, OKR Drift Detector, Board-Cycle Prep): see the X4 table in managed-agents/README.md.

See managed-agents/README.md for manifest conventions, security tiers, and steering-event examples. Deploy with scripts/deploy-managed-agent.sh, which resolves manifest conveniences before posting to POST /v1/agents using the Managed Agents API.

Vertical plugins

Grouped by where the work sits. Each plugin's practice setup is what tailors it to your team — start there.

Craft & delivery

Corporate & portfolio strategy

Measurement frameworks

If you run both BSC and OKR: a common pattern is BSC setting multi-year strategic themes and the causal map, with OKR cascading quarterly execution underneath specific BSC objectives. Don't run both cascades on the same objective set independently — pick which plugin owns which layer and record it in both practice profiles.

Platform

Planned, not yet built: thought-leadership.

MCP connectors

These plugins ship connectors for the systems strategy teams live in. A connector gives Claude the ability to read from and (where scoped) write to your data; the skills and commands use them. Skills produce usable output when no connector is configured — connectors are enhancements, not hard dependencies unless documented otherwise.

Per-plugin connector details: each plugin's CONNECTORS.md (contributor index) links to <plugin>/CONNECTORS.md.

Connectors marked "customer subscription" need your own account and API key. Configure them in each plugin's .mcp.json or via claude mcp in Claude Code.

Cold-start interview and practice profile

Every plugin ships a guided practice-setup interview — the cold-start path that learns your playbook and writes two layers of profile:

Run once per plugin you install: /<plugin>:practice-setup (or --quick for 2 minutes, --full for seed-document review). The interview shows a summary and writes only after you confirm. Framework: consulting/references/practice-setup-framework.md.

Living profile. Every other skill uses propose profile update — show the exact diff, ask, write on yes. No skill except practice-setup auto-writes a full profile.

Making it yours

These are reference templates. They get better when you tune them to how your team works — and the customization mechanism is the plugin itself, not a config file buried in a repo.

• Run the practice setup. It is the customization mechanism. It asks how your practice works, reads your seed documents, and writes your practice profile after you confirm the summary. Every other skill reads from that profile.
• Edit the practice profile. Org-wide facts live at ~/.claude/plugins/config/claude-for-strategy/org-profile.md; plugin-specific conventions at ~/.claude/plugins/config/claude-for-strategy/<plugin>/CLAUDE.md. Edit either file directly for small fixes. They survive plugin updates.
• Propose profile updates from any skill. When a stable convention surfaces during work (repeated tone corrections, a new RAID threshold, tracker column names), skills show the exact profile change and ask before writing. Only practice-setup auto-applies a full profile write — everything else proposes first.
• Re-run setup. /<plugin>:practice-setup again for a full re-interview when your practice shifts materially.
• Swap connectors. Point .mcp.json at your project tracker, slides tool, observability platform, and data sources. Skills fall back gracefully when a connector isn't configured.
• Bring your playbook and templates. Drop your terminology, house style, and branded templates into the practice profile and references/. The skills will pick them up.
• Fork skills for house style. Every skill is a markdown file under skills/. Edit the steps, the gates, the output format.
• Add watchers. Scheduled agents live under <plugin>/agents/ with their system prompt and cadence. Job-style names for on-demand work map to slash commands in each plugin's README — not to duplicate files under agents/.

No build step. Everything is markdown and JSON.

Skill & command reference

The full map across all plugins. The practice setup is the first thing to run in any plugin.

consulting

corporate-strategy

market-intelligence

transformation