How agent runs execute, and where authority comes from
Bundle OKF 0.2 · 0 conceitos · pulse-ops-ai/secure-home-agent-platform
Open source Repository Open in the app JSON
About
# How agent runs execute, and where authority comes from
Portable reasoning context for an agent running on this platform. It answers
questions you may need to answer *while running*: what you are allowed to do,
what a refusal means, and what you may truthfully report about what happened.
This is **context, not authority**. Nothing here grants a capability, authorizes
an action, or overrides live state. Where this bundle and the running system
disagree, the running system is right and this bundle is stale.
| Concept | Read |
|---|---|
| where authority comes from, and what may run next | [authority-and-lifecycle.md](authority-and-lifecycle.md) |
| how a single effect is bounded, and what a timeout means | [effect-boundary.md](effect-boundary.md) |
| which identity proves what, and when a write is real | [identity-and-finalization.md](identity-and-finalization.md) |
Details
- Kind
- OKF bundles
- Topic
- AI, RAG & memory
- Publisher
- pulse-ops-ai
- Origin
- okf_github
- Category
- dados
- Version
- 0.2
- Added
- 2026-09-08 16:04:09
- Updated
- 2026-09-08 16:04:09
- Origin id
pulse-ops-ai/secure-home-agent-platform:knowledge/platform/runner-model/index.md