Agent skill

Decision Register Builder

by pnp in pnp/sharepoint-skills

Analyze SharePoint project content to identify, classify, consolidate, and prepare evidence-based project decision register entries.

MITAuto-check passedDocuments & Office

Install Decision Register Builder

skills CLI
$ npx skills add pnp/sharepoint-skills --skill decision-register-builder -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install pnp/sharepoint-skills decision-register-builder --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/pnp/sharepoint-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/Skills/decision-register-builder/decision-register-builder .claude/skills/decision-register-builder && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
decision-register-builder
GitHub stars
131
Token cost
~3.4k tokens
SKILL.md length
1,726 words
Files
1
Skills in repo
51
Repo updated
First seen
Licence
MIT

At a glance

Analyze SharePoint project content to identify, classify, consolidate, and prepare evidence-based project decision register entries.

  • Works in 10 steps: Discover and Establish Existing Register… → Extract Decision Evidence → Classify Candidates → …
  • The user asks to build
  • SKILL.md covers Purpose, Before Starting, Preferred Sources and Decision Status Model, plus 16 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Decision Register Builder is an agent skill from pnp/sharepoint-skills. Analyze SharePoint project content to identify, classify, consolidate, and prepare evidence-based project decision register entries. Use when the user asks to build, update, review, extract, or prepare a decision register or decision log from project documents, meeting minutes, status reports, notes, or existing SharePoint project content.

Its SKILL.md is about 3.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Documents & Office, covering Cloud office suites, Architecture decision records and Meeting notes and agendas. It works with Microsoft SharePoint. The repository describes itself as: Skills for Copilot in SharePoint. The licence is MIT.

When your agent uses it

  • The user asks to build
  • Prepare a decision register
  • Decision log from project documents
  • Meeting minutes

Example prompts

  • “/decision-register-builder”

Workflow steps

10 steps, taken from the step headings in SKILL.md.

  1. Discover and Establish Existing Register State
  2. Extract Decision Evidence
  3. Classify Candidates
  4. Consolidate and Deduplicate
  5. Resolve Conflicting Evidence
  6. Relate Decisions
  7. Review Register Quality
  8. Produce Review Result
  9. Prepare Register Entries
  10. Optional Standalone HTML Report

What it can do on your machine

Read from SKILL.md and the folder at commit aa9eb14. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Decision Register Builder loads about 3.4k tokens when it runs. Until then it costs about 92 tokens; SKILL.md has 1,726 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~92
When it runs · the whole SKILL.md, loaded when a task matches
~3.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from pnp/sharepoint-skills at commit aa9eb14, republished under its MIT licence (© pnp). 1,726 words, ~3,418 tokens.

Download SKILL.mdSave it as .claude/skills/decision-register-builder/SKILL.md (or your agent's skills folder).
name
decision-register-builder
description
Analyze SharePoint project content to identify, classify, consolidate, and prepare evidence-based project decision register entries. Use when the user asks to build, update, review, extract, or prepare a decision register or decision log from project documents, meeting minutes, status reports, notes, or existing SharePoint project content.

Decision Register Builder

Purpose

Build or review a project Decision Register from evidence available in SharePoint. Identify explicit decisions and genuine decision requirements, separate them from discussion/actions, consolidate duplicate references, and prepare structured register entries. Never invent project facts.

Before Starting

  1. Determine the project/content scope from the prompt, selected files, current SharePoint location, or clearly related project content.
  2. Treat the current document library or selected files as the starting context, not automatically as the complete project scope.
  3. Discover relevant structured SharePoint lists on the current project site before concluding that a project register does not exist.
  4. In particular, look for an existing Decision Register implemented as a SharePoint list. Common names include Decision Register, Decision Log, Decisions, Project Decisions, and reasonable project-specific equivalents.
  5. If a likely Decision Register list exists, inspect its schema and current entries before proposing new decision IDs or concluding that no existing register is available.
  6. Prefer current authoritative artifacts.
  7. Treat an existing Decision Register as authoritative for existing IDs and current registered state unless newer authoritative evidence supersedes it.
  8. Use only content available to the current user.
  9. Ask for clarification only when project scope is materially ambiguous. Do not ask merely because the register uses a different but recognizable list name.

Preferred Sources

Steering/project/architecture board minutes, meeting notes, status reports, charter, solution/architecture documents, risk/issue/action registers, existing Decision Register, change-control records, and other project evidence containing approvals, rejections, deferrals, selections, or unresolved choices. Do not assume every source type exists.

Decision Status Model

  • Open — a real decision is required but no final choice is approved.
  • Proposed — a specific choice is proposed but awaits approval.
  • Decided — authoritative evidence records an approved/accepted/selected/rejected/finalized choice.
  • Deferred — authoritative evidence explicitly postpones the decision.
  • Superseded — a previously decided item was replaced by a newer decision.
  • Unknown — a decision item exists but its current state cannot be established reliably.

Do not infer Decided because an option is preferred, recommended, discussed, planned, or assigned for investigation.

What Counts as a Decision

A Decided item requires evidence of an actual choice or authoritative disposition. Strong indicators include approved, agreed, decided, selected, accepted, rejected, confirmed, authorized, adopted, or equivalent explicit language.

An Open/Proposed candidate requires a concrete choice, approval, or unresolved question that materially affects the project.

Examples:

  • "The Architecture Board must select the production hosting topology." → Open
  • "The team recommends Azure hosting, subject to Architecture Board approval." → Proposed
  • "The Architecture Board approved Azure hosting for production." → Decided

What Does Not Count

Do not create a decision entry for ordinary discussion, informational statements, routine actions/tasks, risks/issues by themselves, agenda topics, recommendations needing no approval, assumptions not treated as decisions, or unauthorized opinions.

"The Architecture Board will review the hosting options next Thursday" is not evidence of a Decided item. It may support an existing Open decision.

Register Fields

Capture when supported:

  • Decision ID
  • Title
  • Status
  • Decision / Required Decision
  • Decision Owner
  • Decision Date
  • Required By
  • Context
  • Rationale
  • Alternatives
  • Impact
  • Related Items
  • Evidence
  • Confidence

Use Not stated when an important output field is unsupported. Never fill gaps with plausible assumptions.

Step 1: Discover and Establish Existing Register State

1.1 Discover structured project lists

Before extracting new decision candidates, inspect the current project site for relevant structured SharePoint lists.

Do not conclude that no Decision Register exists merely because it is absent from the current document library, selected files, or current folder. A project Decision Register is commonly implemented as a SharePoint list rather than as a document.

Search for likely register lists using:

  • list title;
  • list description, when available;
  • recognizable decision-register columns;
  • project context.

Common list names include:

  • Decision Register
  • Decision Log
  • Decisions
  • Project Decisions
  • reasonable project-specific equivalents

A differently named list can still be the Decision Register when its schema clearly represents project decisions.

Useful schema indicators include columns equivalent to:

  • Decision ID
  • Title
  • Status
  • Decision / Required Decision
  • Decision Owner
  • Decision Date
  • Required By
  • Rationale
  • Alternatives
  • Impact
  • Related Items
  • Evidence

Do not require every column to exist.

1.2 Select the authoritative register

When one likely Decision Register is found:

  1. Read its existing entries before proposing new entries.
  2. Treat it as the authoritative register for existing Decision IDs and registered state.
  3. Preserve existing IDs and never renumber them.
  4. Match documentary evidence against existing entries before creating candidates.
  5. Detect updates, confirmations, deferrals, superseding decisions, and duplicates.
  6. Determine the highest numeric ID only when useful for proposing a future ID.

When multiple plausible decision lists are found:

  1. Prefer the list clearly associated with the current project.
  2. Prefer the list whose schema and contents best match a project Decision Register.
  3. Do not silently combine unrelated decision lists.
  4. If two or more lists remain materially ambiguous, state the ambiguity and ask the user which register is authoritative before proposing updates.
1.3 Handle inaccessible or missing registers

Distinguish these cases:

  • Register found and readable: use it.
  • Likely register found but inaccessible: state that it appears to exist but could not be read with the current user's access. Do not say that no register exists.
  • No likely register found after site-level list discovery: state that no accessible Decision Register was found on the current project site and use temporary candidate IDs.
  • Structured-list discovery is unavailable in the execution environment: explicitly state that the existing register could not be verified. Do not claim that no register exists.

Only after site-level structured-list discovery has been attempted may the skill fall back to NEW-01, NEW-02, etc.

If no accessible register can be established, use temporary candidate IDs NEW-01, NEW-02, etc. unless the user provides an ID convention.

Step 2: Extract Decision Evidence

Identify passages indicating a decision was made; approval/rejection occurred; a decision is required; a proposed choice awaits approval; a decision was deferred; a prior decision was replaced; or owner/deadline/rationale/alternatives/impact changed. Keep evidence tied to its source.

Step 3: Classify Candidates

For every candidate:

  1. Verify it is a genuine decision item.
  2. Assign the most evidence-supported status.
  3. Match it to an existing entry if applicable.
  4. Capture only supported fields.
  5. Assign confidence:
    • High — explicit decision language and clear authority/context.
    • Medium — decision intent is clear but a material aspect is ambiguous.
    • Low — possible decision, but evidence is insufficient for safe registration.

Never present Low-confidence candidates as confirmed decisions.

Show full SKILL.md (684 more words)Show less

Step 4: Consolidate and Deduplicate

Treat references as the same decision when subject, choice, context, and project impact materially match. Merge evidence, prefer the newest authoritative evidence for current status, preserve useful history, and do not create duplicates merely because wording differs. Do not merge separate decisions merely because they concern the same workstream.

Step 5: Resolve Conflicting Evidence

  1. Prefer the source authoritative for the decision.
  2. Prefer newer authoritative evidence over older evidence.
  3. Never silently reconcile material contradictions.
  4. State material conflicts.
  5. Use Unknown if current state cannot be established safely.
  6. If newer evidence explicitly replaces an earlier decision, mark the earlier decision Superseded and link entries when possible.

Example: status report says "Azure hosting recommended"; Architecture Board minutes say "decision deferred pending cost comparison" → current status is Deferred.

Step 6: Relate Decisions

When explicitly supported, connect decisions to actions, risks, issues, milestones, deliverables, dependencies, and other decisions. Never invent relationships.

Step 7: Review Register Quality

When a register exists, identify material issues such as:

  • unresolved decisions past Required By;
  • missing owner where ownership is expected;
  • decided entries missing an available decision date;
  • duplicates;
  • conflicting status;
  • superseded decisions not updated.

Do not flag optional metadata merely because it is blank.

Step 8: Produce Review Result

1. Decision Register Summary

State:

  • Decision Register discovery result, including the SharePoint list name when found;
  • existing decisions reviewed;
  • new decision candidates;
  • existing entries requiring update;
  • low-confidence candidates requiring human review.

If no register was used, explain whether no accessible register was found after site-level discovery, a likely register was inaccessible, or structured-list discovery was unavailable.

2. New Decision Candidates

For each: Candidate ID, Title, Proposed Status, Decision/Required Decision, Owner, Decision Date or Required By, Impact, Related Items, Evidence, Confidence.

3. Existing Entries to Update

For each: Decision ID, current registered value/status, evidence-supported update, reason, evidence, confidence.

4. Items Requiring Human Review

Separate ambiguous/conflicting candidates that should not be registered automatically and explain what is unclear.

5. Register Quality Findings

Report only material evidence-supported findings. If none, say so.

Step 9: Prepare Register Entries

When asked, produce structured entries suitable for a SharePoint list. Do not modify a SharePoint Decision Register unless the user explicitly asks, the environment supports it, and the user has permission. Never claim a write succeeded unless it actually did.

Step 10: Optional Standalone HTML Report

When requested, generate a complete standalone HTML5 webpage from the validated review.

Requirements:

  • <!DOCTYPE html>, <html>, <head>, UTF-8 charset, viewport, title, embedded CSS, <body>;
  • no JavaScript, external CSS/fonts/images/CDNs, or required network access;
  • directly browser-openable and print-friendly;
  • include project/reporting date, summary, new candidates, updates, human review, quality findings, evidence/coverage, confidence;
  • HTML-encode project-derived content;
  • do not reinterpret the validated result while formatting.

When file creation is supported, prefer: Decision-Register-Review-[Project-Name]-[YYYY-MM-DD].html

Do not claim a file was saved unless it actually was.

Evidence Rules

Evidence first. Never fabricate facts. Prefer authoritative sources for the facts they own and newer authoritative evidence when state changes. Preserve traceability. Do not turn recommendations/actions/risks into decisions without explicit decision evidence. Do not infer authority solely from job titles, rationale from outcomes, or alternatives from technical possibilities. Signal uncertainty.

Date Rules

  • Open/Proposed/Deferred with Required By before reporting date → overdue.
  • Due on reporting date → due today, not overdue.
  • Future → not overdue.
  • Calculate lateness from Required By, not Decision Date.
  • If reporting date is absent, use the current date available to the environment.

Validation Checklist

  • Every Decided item has evidence of an actual authoritative choice.
  • Recommendations/discussions were not converted into decisions.
  • Site-level structured SharePoint lists were checked before concluding that no Decision Register exists.
  • The current document library was not treated as the complete project scope when relevant site lists could exist.
  • A likely but inaccessible register was not reported as nonexistent.
  • Existing IDs were preserved.
  • New candidates were deduplicated against existing entries.
  • Current status uses newest authoritative evidence.
  • Material conflicts are disclosed.
  • Superseded decisions are linked when supported.
  • Unknown fields were not invented.
  • Related items are evidence-supported.
  • Due today is not called overdue.
  • Low-confidence candidates are separated for human review.
  • Proposed changes are distinguished from actual SharePoint writes.
  • Requested HTML is standalone and dependency-free.

© pnp, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in Skills/decision-register-builder/decision-register-builder of pnp/sharepoint-skills.

Open the folder on GitHubat commit aa9eb14

Compare with similar skills

Decision Register Builder next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Decision Register Builder compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Decision Register Builder this skillpnp/sharepoint-skills131—~3.4kAutomated safety check: PassMIT
Msgraphcodemie-ai/codemie-code294—~4.1kAutomated safety check: PassApache-2.0
aai-cli Microsoft 365aai-labs/agent-barn109—~1.2kAutomated safety check: PassApache-2.0
Colleague DistillationZhixiangLuo/10xProductivity478—~2.1kAutomated safety check: NotesMIT
Workiqmicrosoft/work-iq1k—~15kAutomated safety check: PassCustom licence
Workiq Previewmicrosoft/work-iq1k—~3.3kAutomated safety check: PassCustom licence

Similar skills

  • Msgraph

    codemie-ai/codemie-code

    Work with Microsoft 365 services via the Graph API — emails, calendar events, SharePoint sites (read and write), Teams chats and channel messages, OneDrive files, OneNote notebooks, Planner task…

    294 GitHub stars~4.1k tokensUpdated yesterday
    Documents & OfficeAuto-check passed
  • aai-cli Microsoft 365

    aai-labs/agent-barn

    Guides work with Outlook, OneDrive, SharePoint, Teams, Excel, To Do and Planner through aai-cli's Microsoft Graph commands, starting from which service owns the data.

    109 GitHub stars~1.2k tokensUpdated today
    Documents & OfficeAuto-check passed
  • Colleague Distillation

    ZhixiangLuo/10xProductivity

    Distill a colleague into a reusable AI skill (work + persona) using tool connections — Slack, Slack AI, Jira, GHE, Bitbucket, Confluence, SharePoint, Teams, Outlook, Notion, Linear, Google Docs, and…

    478 GitHub stars~2.1k tokensUpdated 2 mo ago
    Documents & OfficeAuto-check: notes
  • Workiq

    microsoft/work-iq

    Official

    WorkIQ tools for Microsoft 365 workplace data and actions. An agent skill from microsoft/work-iq.

    1k GitHub stars~15k tokensUpdated yesterday
    Documents & OfficeAuto-check passed
  • Workiq Preview

    microsoft/work-iq

    Official

    WorkIQ tools for Microsoft 365 workplace data and actions. An agent skill from microsoft/work-iq.

    1k GitHub stars~3.3k tokensUpdated yesterday
    Documents & OfficeAuto-check passed
  • Hunt Ntlm Info

    sickn33/agentic-awesome-skills

    Hunt NTLM/Negotiate information disclosure on internet-reachable IIS/SharePoint/Exchange.

    47k GitHub starsUsed in 1 repo~4.7k tokens
    Documents & OfficeAuto-check passed

More from pnp/sharepoint-skills

All 51 skills in this repo
  • Scorecard Matrix

    pnp/sharepoint-skills

    Generates a polished, self-contained HTML heatmap scorecard — a weighted comparison matrix where entities (rows) are scored across dimensions (columns), with computed totals, rank badges, and a…

    131 GitHub stars~1.9k tokensUpdated 2 days ago
    Auto-check passed
  • Analyze Document Library

    pnp/sharepoint-skills

    Analyze the current SharePoint document library in read-only mode and produce a structured summary of files, folders, file types, recent activity, naming issues, and organization recommendations.

    131 GitHub stars~899 tokensUpdated 2 days ago
    Auto-check passed
  • Broken Link Auditor

    pnp/sharepoint-skills

    Audits SharePoint pages, news posts, and hyperlink fields for broken or risky links and saves a self-contained HTML link-health report to the site.

    131 GitHub stars~2.5k tokensUpdated 2 days ago
    Auto-check passed
  • Custom Image Tagger

    pnp/sharepoint-skills

    Analyze selected construction images, create missing object metadata columns, and write concise visual metadata back to SharePoint columns using explicit image-analysis, list-schema, list-update…

    131 GitHub stars~1.2k tokensUpdated 2 days ago
    Auto-check passed
  • Dossier

    pnp/sharepoint-skills

    Renders a polished, self-contained HTML briefing from any data source — SharePoint lists, uploaded documents, or a verbal description.

    131 GitHub stars~1.9k tokensUpdated 2 days ago
    Auto-check passed
  • Exec Report

    pnp/sharepoint-skills

    Generates a polished, self-contained HTML executive report or dashboard from any data source — SharePoint lists, CSV exports, or a user description.

    131 GitHub stars~2k tokensUpdated 2 days ago
    Auto-check passed

Questions about Decision Register Builder

What does Decision Register Builder do?

Analyze SharePoint project content to identify, classify, consolidate, and prepare evidence-based project decision register entries. Decision Register Builder is an agent skill from pnp/sharepoint-skills. Analyze SharePoint project content to identify, classify, consolidate, and prepare evidence-based project decision register entries.

When should I use Decision Register Builder?

Decision Register Builder fits situations like: the user asks to build; prepare a decision register; decision log from project documents; meeting minutes.

How do I install Decision Register Builder in Claude Code?

Run `npx skills add pnp/sharepoint-skills --skill decision-register-builder -a claude-code`. Or copy the skill folder (Skills/decision-register-builder/decision-register-builder in pnp/sharepoint-skills) into .claude/skills/decision-register-builder in your project. Claude Code loads it when a task matches its description.

How do I install Decision Register Builder in Codex?

Run `npx skills add pnp/sharepoint-skills --skill decision-register-builder -a codex`. Or copy the skill folder (Skills/decision-register-builder/decision-register-builder in pnp/sharepoint-skills) into .agents/skills/decision-register-builder in your project. Codex loads it when a task matches its description.

Can I use Decision Register Builder in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add pnp/sharepoint-skills --skill decision-register-builder -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/decision-register-builder, .gemini/skills/decision-register-builder, .github/skills/decision-register-builder and .opencode/skills/decision-register-builder in your project.

What does Decision Register Builder need to run?

SKILL.md names no scripts, command-line tools or credentials: Decision Register Builder is instructions for the agent only.

Does Decision Register Builder access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Decision Register Builder safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Decision Register Builder use?

Decision Register Builder is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Decision Register Builder use?

About 3.4k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Decision Register Builder?

Skills that share tags, products or a category with Decision Register Builder: Msgraph (codemie-ai/codemie-code, 294 stars), aai-cli Microsoft 365 (aai-labs/agent-barn, 109 stars), Colleague Distillation (ZhixiangLuo/10xProductivity, 478 stars) and Workiq (microsoft/work-iq, 1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Decision Register Builder?

pnp (a GitHub organization) maintains it in pnp/sharepoint-skills, which has 131 GitHub stars. The repository holds 51 skills in this directory. The repository was last updated on October 6, 2026.

Source: pnp/sharepoint-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.