Mixed freshness. Depending on the panel, rows here are read live from elisa-core or come from a frozen copy of ELISA’s local store taken on 22 October 2026 — each panel says which, underneath itself. No agent is running.

ELISA
Connecting

Agent 5

Review code, dependencies, GitHub Actions and agent/tool actions for vulnerabilities and produce findings reports; it holds the defensive review family and no offensive tooling.

Agent 5

Not connectedIdle — nothing queued

Security Review (defensive) · Engineering (not in the ELISA spec)

engineering.agent_5

Profile charter name: Suraksha — a label from the Hermes profile. Nothing routes on it; routing is by id only.

Measured — state right now

Idle — nothing queued. A profile exists, its queue is empty and nothing is pending. Idle describes the queue, not a live process. This UI is not connected to a running agent and never starts one.

Derived state
Idle — nothing queued
Backend reason
no open job, no open card, nothing pending
Runtime attachment
Not connected — this UI holds no fleet runtime client, so it cannot observe liveness.
Hermes profile
engineering-a5
Profile built
Yes
Runtime lock
Not locked · sector dispatchable: yes
May create its own cards
No
Spec source
specs/production.json
Derived 5 days ago · STALE, this reading may no longer be true · 2026-10-20T12:45:44.000Zfrom elisa.db:agents.runtime_locked+sectors.runnable (via dispatchable()) · elisa.db:approvals.status='pending' · elisa.db:jobs.status · kanban.db:tasks.status (assignee=profile)

Everything below is this agent’s declared specification, read from elisa.db. It is what the agent is defined to be able to do — not evidence that it has run a job, used a skill or produced an output.

Mission (declared)

Review code, dependencies, GitHub Actions and agent/tool actions for vulnerabilities and produce findings reports; it holds the defensive review family and no offensive tooling.

Will not (declared, and enforced at the gate)

  • run, launch or invoke any offensive/exploitation tool, scanner or payload (metasploit, sqlmap, burp, nmap-style recon, injection/fuzzing) - defensive review only, read and report
  • modify application code to apply a fix itself; it files findings and recommendations for another agent to implement
  • write into C:/elisa/engine/** (L-0001) or C:/elisa/ui/** (L-0002)

Owns (record types)

  • security_review_report
  • vulnerability_finding
  • remediation_recommendation

Skills (declared)

  • agentic-actions-auditor
  • gha-security-review
  • secure-code-guardian
  • security-auditor
  • security-reviewer

Skills deliberately withheld

  • metasploit-framework — offensive tooling: legitimate for Deepak by hand, but an unattended autonomous profile must not hold exploitation frameworks - unassigned to any agent
  • sqlmap-database-pentesting — offensive DB exploitation; unassigned to any agent, available to him manually
  • burp-suite-testing — offensive web exploitation tooling; unassigned to any agent
  • red-team-tools — offensive; unassigned to any agent
  • red-team-tactics — offensive; unassigned to any agent
  • ethical-hacking-methodology — offensive end-to-end methodology; unassigned to any agent
  • active-directory-attacks — offensive; unassigned to any agent
  • linux-privilege-escalation — offensive privilege escalation; unassigned to any agent
  • windows-privilege-escalation — offensive privilege escalation; unassigned to any agent
  • privilege-escalation-methods — offensive; unassigned to any agent
  • sql-injection-testing — offensive injection; unassigned to any agent
  • xss-html-injection — offensive injection; unassigned to any agent
  • html-injection-testing — offensive injection; unassigned to any agent
  • idor-testing — offensive; unassigned to any agent
  • api-fuzzing-bug-bounty — offensive fuzzing; unassigned to any agent
  • shodan-reconnaissance — offensive recon; unassigned to any agent
  • ssh-penetration-testing — offensive; unassigned to any agent
  • metasploit-framework-and-the-rest — the full ~45-skill offensive-security family is deliberately assigned to NO agent; it stays available to Deepak by hand and never rides on an autonomous profile

Jobs (declared job types)

No job types declared.

MCP requirements

  • microsoft-learnreadDeclared, not wiredmicrosoft_docs_search

Persona (declared)

Communication
see this profile's SOUL.md
Reasoning
see charter.json -> contract and spec.will_not
Pushback
run, launch or invoke any offensive/exploitation tool, scanner or payload (metasploit, sqlmap, burp, nmap-style recon, injection/fuzzing) - defensive review only, read and report

Inputs (reads)

  • security_review_request
  • backend_code_change
  • ci_config_draft

Outputs (produces)

  • security_review_report
  • vulnerability_finding
  • remediation_recommendation

Routing

Escalates to
engineering.orchestrator
Hands off to
engineering.agent_1engineering.agent_4engineering.orchestrator

Approval policy

Draft only — produces drafts, never acts without a human review

Idle missions

None. This agent does nothing unprompted — there is no idle mission to run.