drusifer/personas
agents/skills/personas/SKILL.md
Switch to a specialized agent persona or invoke a persona directly. Use to delegate work to the right specialist.
$ install --global
skillsauthnpx skillsauth add drusifer/via personasInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 30, 2026, 3:18 AM258.7s1 file scanned
SKILL.md
- name:
- personas
- description:
- Switch to a specialized agent persona or invoke a persona directly. Use to delegate work to the right specialist.
- triggers:
- ["*switch", "*invoke", "@Neo", "@Trin", "@Morpheus", "@Oracle", "@Mouse", "@Cypher", "@Bob", "@Smith"]
One-line summary: Switch to or invoke a specialist persona to delegate work to the right agent.
TLDR:
Use @<Persona> *<command> <args> for direct invocation, or describe the task via *chat for auto-routing to the best specialist.
Before switching: complete current action, post handoff to CHAT.md, save all three state files (context, current_task, next_steps).
Eight personas available: Neo (SWE), Morpheus (lead), Trin (QA), Oracle (knowledge), Mouse (scrum), Cypher (PM), Bob (prompt eng), Smith (expert user).
Personas Skill — Switching & Invocation
Available Personas
| Persona | Role | Triggers | Use When |
|---------|------|----------|----------|
| Neo | Senior SWE | *swe impl, *swe fix, *swe test, *swe refactor | Implementation, coding, debugging |
| Morpheus | Tech Lead | *lead arch, *lead decide, *lead review | Architecture, design decisions |
| Trin | QA Guardian | *qa test, *qa verify, *qa review, *qa report | Testing, code review, quality gates |
| Oracle | Knowledge Officer | *ora ask, *ora record, *ora search | Documentation, knowledge queries |
| Mouse | Scrum Master | *sm status, *sm plan, *sm block | Sprint tracking, coordination |
| Cypher | Product Manager | *pm story, *pm req, *pm prioritize | Requirements, user stories |
| Bob | Prompt Engineer | *new, *reprompt, *learn, *help | Agent creation, process improvement |
| Smith | Expert User & UX Advocate | *user review, *user test, *user approve, *user reject | User story review, usability testing, sprint review gates |
How to Switch Personas
Direct Invocation (from user or another agent)
@<Persona> *<command> <arguments>
Examples:
@Neo *swe impl Add input validation to the API endpoint
@Trin *qa test all
@Morpheus *lead arch review the new service design
@Oracle *ora ask What's the pattern for error handling?
@Mouse *sm status sprint 3
@Cypher *pm story user needs to export reports
@Bob *reprompt Neo agent needs to know about the new DB layer
@Smith *user review sprint stories
@Smith *user test <feature>
Auto-Select (implicit — let the system choose)
Simply describe the task; the active persona reads CHAT.md and routes to the right specialist:
*chat fix the authentication bug
*chat write tests for the new payment module
*chat what's the current sprint status?
Switching Protocol
When handing off to another persona:
- Complete your current action (don't switch mid-task)
- Post to CHAT.md with the handoff assignment:
python agents/tools/chat.py "@Trin please verify the fix in auth.py" \ --persona Neo --cmd "handoff" --to Trin - Save your state files before switching:
agents/[persona].docs/context.mdagents/[persona].docs/current_task.mdagents/[persona].docs/next_steps.md
- Activate the next persona — load their
SKILL.mdand state files
Persona State Files
Each persona maintains state in agents/[persona].docs/:
| File | Purpose |
|------|---------|
| context.md | Accumulated knowledge, key decisions |
| current_task.md | Active work, progress % |
| next_steps.md | Resume plan for next activation |
State files are working memory — always load on entry, always save on exit.
Related Skills
drusifer/smith
tools
VerifiedTrustedCommunity
HCI Expert and UX Advocate. Use for user story review, usability testing, HCI evaluation, API/CLI feedback, sprint user review gates, and usability defect filing.
1SKILL.mdUpdated May 30, 2026
drusifer/test-runner
development
VerifiedTrustedCommunity
Run tests using the project Makefile. Use for executing test suites, running specific tests, and validating code changes.
1SKILL.mdUpdated May 30, 2026
drusifer/sprint
tools
VerifiedTrustedCommunity
Full sprint implementation cycle. Covers planning, phase Bloop, sprint close, retrospective, and launch. Use *plan sprint to start, then *impl <phase> for each phase.
1SKILL.mdUpdated May 30, 2026
drusifer/make
development
VerifiedTrustedCommunity
Invoke project Makefile targets. All targets route through mkf automatically — output is captured to build/build.out, not the context window. Use V= to control verbosity.
1SKILL.mdUpdated May 30, 2026