phere-9-agent-manager-e35296 — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited phere-9-agent-manager-e35296 (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use this skill to coordinate developer-side work on Phere with the named Phere 9-Agent Manager operating model. The name stays "9-Agent" for the core team, but enterprise mode adds specialist and optional business agents behind strict gates. This is an enterprise-style operating model, not a claim that the app is certified, deployed, or accepted by any company.
orchestrator for broad or unclear work.references/PHERE_9_AGENT_ROSTER.md for role cards.references/PHERE_AGENT_ROUTING_ALGORITHM.md before routing large or ambiguous work.references/PHERE_ENTERPRISE_AGENT_OPERATING_MODEL.md for heavy-project coordination.references/PHERE_AGENT_TRUTHFULNESS_RULES.md before making status, capability, production, dependency, or business claims.references/PHERE_AGENT_QUALITY_CONTRACTS.md when defining done criteria.references/PHERE_9_AGENT_WORKFLOWS.md for task routing and handoff rules.references/PHERE_PRODUCTION_SECURITY_GATES.md before release, deployment, database, auth, or security-sensitive work.references/PHERE_OPTIONAL_BUSINESS_AGENTS.md only after coding is done or when the user explicitly asks for marketing, sales, launch, support, pricing, or growth work.references/PHERE_CODEX_HANDOFF_QUEUE_PLAN.md when the user asks to use the already-open Codex app, a no-cost handoff queue, or current-Codex manual agents.references/PHERE_FREE_LOCAL_AI_WORKER_PLAN.md only when the user asks to build free/local AI-code-writing background workers.Only spawn real subagents when the user explicitly asks for agent delegation, parallel agent work, or to use the agent team. Otherwise, act as the selected role yourself and state which role you are using.
package.json for dependency versions and current command output for validation status.Use the local helper when a concrete brief or gate is useful:
npm run agents:list
npm run agents:brief -- planner "Plan the next stability sprint"
npm run agents:route -- "Add vendor payment reminders"
npm run agents:gate
npm run agents:truth
npm run agents:start -- "Improve Saman performance"
npm run agents:status
npm run agents:activity
npm run agents:activity -- tester
npm run agents:flow
npm run agents:eta
npm run agents:handoff -- "Saman performance and mobile verification"
npm run agents:handoff-next
npm run agents:handoff-status
npm run agents:handoff-done -- frontend "Saman UI reviewed"
npm run agents:run -- "Saman performance and mobile verification"
npm run agents:logs
npm run agents:stop
npm run agents:update -- "tester preparing mobile QA"
npm run agents:update -- tester "preparing mobile QA checklist"
npm run agents:done -- "status handoff complete"End agent work with:
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.