Poesis
Usage / Use case Partial
Generate artifacts through declared workflows
Produce a grounded document or software artifact through assigned roles, bounded workflow execution and explicit deliverable contracts, then obtain the required revision-bound review before advancement. Technical checks, substantive acceptance and authorized deployment remain separate decisions.
Delivery agent / Operational knowledge owner / Head of AI / Platform / Ops / SRE lead
Platform
What serves this use case
Affordances
- Bounded generative delivery Partial
Values
Values composed with this use case
- Value Partial
Distinguish generated output from accepted completion
Inspect binding resolution, grounding constraints, missing inputs and job outcomes before accepting or reusing a generated deliverable.
Supported by Affordances(1)
Supported by Capabilities(1)
Supported by Features(1)
Realized through Use cases(6)
- Value Partial
Delegate bounded work with evidence and recourse
Build toward delegated workflows with explicit obligations, decision boundaries and controlled effects, retaining evidence for intervention and recovery.
Supported by Affordances(1)
Supported by Capabilities(3)
Supported by Features(2)
Realized through Use cases(7)
- Approve an agent resource and action mandate
- Execute a bounded response and inspect the receipt
- Deliver a reviewed artifact through a local workflow
- Run a Mechanism inside declared bounds
- Check a produced artifact against its contract
- Gate an advancement on an authenticated human decision
- Generate artifacts through declared workflows
- Value Partial
Make agentic work reviewable.
Combine role contracts, workflow decisions and artifact evidence, with planned authenticated human gates bound to exact revisions.
Supported by Affordances(1)
Supported by Capabilities(1)
Supported by Features(2)
Pains
Pain points
- Pain point Partial
Incorrect bindings, unsupported narrative and partial files pass for completed deliverables.
Reviews discover missing evidence and output errors only after reports have been distributed.
Addressed by Affordances(1)
Addressed by Capabilities(3)
Addressed by Features(4)
Experienced by Actor types(6)
Addressed through Use cases(13)
- Author and maintain reusable output bindings
- Preview bindings before generating an artifact
- Set grounding constraints for generated narrative
- Operate recurring report production
- Investigate partial or failed generation
- Deliver a reviewed artifact through a local workflow
- Validate a mediated artifact write
- Define a deliverable schema and template
- Publish artifacts through configured CI checks
- Produce a named IT document family from pinned definitions
- Check a produced artifact against its contract
- Gate an advancement on an authenticated human decision
- Generate artifacts through declared workflows
- Pain point Partial
Every deliverable is hand-crafted, again and again.
ADR packs, evidence, baselines, and roadmaps are rebuilt by hand — and stale against each other by the time they ship.
Addressed by Affordances(1)
Addressed by Capabilities(2)
Addressed by Features(2)
Experienced by Actor types(7)
Addressed through Use cases(8)
- Author and maintain reusable output bindings
- Operate recurring report production
- Generate and review an architecture decision pack
- Reuse captured delivery evidence as definition proposals
- Extract definition candidates from delivery history
- Produce a named IT document family from pinned definitions
- Turn delivery history into definition candidates
- Generate artifacts through declared workflows
- Pain point Partial
AI agents ship work nobody scoped, gated, or can replay.
Agentic delivery state is trapped in chat sessions; pull requests appear without a mandate; there is no event log and no human authority at the layer where it matters.
Addressed by Affordances(1)
Addressed by Capabilities(4)
Addressed by Features(7)
Experienced by Actor types(5)
Addressed through Use cases(14)
- Approve an agent resource and action mandate
- Deliver a reviewed artifact through a local workflow
- Set up a workspace and publish checked artifacts
- Resolve the next eligible workflow step
- Inspect a captured harness invocation sequence
- Assign a delivery responsibility from the role catalog
- Select a practice playbook for a delivery task
- Define a delivery ceremony and its handoffs
- Select versioned instructions for an engineering role
- Run an agentic workflow with deterministic controls
- Adopt the agentic SAFe operating model
- Gate an advancement on an authenticated human decision
- Deliver from a local-first, reproducible plane
- Generate artifacts through declared workflows