From agentic-office
Activate for: workplace memory, person brief, team profile, organisation profile, add person, add project, add terminology, four-layer memory, communication style, stakeholder brief, post-meeting memory revision, who works on, team directory, revise workplace memory, organisation context. NOT for: basic memory CRUD (use official productivity plugin memory-management), cross-context search (use workplace-search).
npx claudepluginhub panaversity/agentfactory-business-plugins --plugin agentic-officeThis skill uses the workspace's default tool permissions.
Read `work.local.md` in the current working directory.
Implements two-tier memory with CLAUDE.md hot cache and memory/ directory to decode acronyms, nicknames, people, projects, and internal terms.
Activate for: search everything, cross-context search, what do we know about, find the decision, search all layers, search across workplace memory, what was agreed, when did we decide, find everything about, who knows about, what have we discussed about, pull everything on, find in workplace memory, search all. NOT for: basic memory lookup (use official productivity plugin memory-management), adding new entries (use workplace-context).
Discovers team structure, terminology, goals, performance frameworks, values, and ways of working by crawling Slack, Notion, Google Drive, Gmail, Calendar; validates with manager before persisting.
Share bugs, ideas, or general feedback.
Read work.local.md in the current working directory.
If it does not exist, tell the user to run /agentic-office:setup first.
TYPE 1: ADD PERSON Input: Name, role, reporting line, communication style, current focus, any sensitivities Output: Formatted person entry for work.local.md Layer 2 + confirmation
PERSON ENTRY FORMAT:
- name: "[Full name]"
role: "[Job title]"
reports_to: "[Manager name]"
communication: >
[How they prefer to receive information; what works; what annoys them]
current_focus: "[What they are working on right now]"
priorities: "[Their current top 1-3 priorities]"
note: "[Anything important about working with this person]"
sensitivity: "[Optional — anything that must not be shared broadly]"
TYPE 2: ADD PROJECT Input: Project name, codename, status, priority, owner, description, current milestone, risks, decisions already made Output: Formatted project entry for work.local.md Layer 3 + confirmation
PROJECT ENTRY FORMAT:
- name: "[Project full name]"
codename: "[Internal codename if different]"
status: "[PLANNING / IN PROGRESS / AT RISK / BLOCKED / COMPLETE]"
priority: "[P1 / P2 / P3]"
owner: "[Named person]"
description: >
[What this project is; why it matters; key context]
current_milestone: "[What is happening now]"
next_milestone: "[What comes next; by when]"
at_risk: "[What could go wrong; what is currently stalled]"
decisions: ["[Decision made; date]"]
key_contacts: ["[Person; role in project]"]
TYPE 3: ADD TERM Input: Term, definition, when to use, when NOT to use, related terms Output: Formatted terminology entry in Layer 4 + confirmation
TERMINOLOGY ENTRY FORMAT:
"[Term]": >
[Definition in plain language — what it means in this organisation]
Use: [When this term is appropriate]
Not: [When NOT to use it — external comms? formal docs? pre-announcement?]
Related: [Synonyms or near-synonyms]
TYPE 4: PERSON BRIEF Input: Person name (or "everyone I'm meeting today" with list) Output: Structured brief per person
PERSON BRIEF FORMAT:
--- [NAME] — [Role] -----------------------------------------------
Reports to: [Name]
Current focus: [What they are working on]
Communication: [How to approach them today]
Today's context: [Anything live that affects this interaction]
How to approach: [Specific guidance for today]
Watch for: [Anything to be aware of]
Do not: [Any specific sensitivities]
-------------------------------------------------------------------
TYPE 5: POST-MEETING UPDATE Input: Meeting name, decisions made, new actions, status changes, new information about people Output: Proposed updates to work.local.md across all relevant layers
PROPOSED UPDATES FORMAT:
MEMORY UPDATE PROPOSAL — [Meeting Name] — [Date]
================================================================
1. UPDATE: [Section]: [What changes]
Current: "[Old text]"
Proposed: "[New text]"
[Continue for each update]
Confirm to apply all? Or specify which updates to apply.
================================================================
TYPE 6: TERMINOLOGY QUERY Input: "What does X mean?" or "What do we call Y?" Output: Load from terminology dictionary in work.local.md; apply in the response; if not found, offer to add
For ADD operations (Types 1-3): Draft the entry, present for confirmation, then write to work.local.md only after user confirms.
For QUERY operations (Types 4, 6): Load from work.local.md and present.
For UPDATE operations (Type 5): Propose all changes, let user select which to apply.
Some entries contain sensitive information:
RULES for sensitive entries:
sensitivity: RESTRICTED flagSTALE MEMORY DETECTION: Flag any entry in work.local.md that has not been updated in >90 days and prompt user to confirm or update.
Common staleness patterns: