Build a reviewable record from the repository while you can still verify it. Package what shipped, expose what is unknown, and stop redrawing the same context for every handover and later question.
A useful result says what evidence supports it, what stayed outside the boundary, and who must confirm the rest.
outputs/…
TWO EQUIVALENT PATHS
A control surface for the job—not another product.
The planned Workbench would make the free capability suite discoverable. Its CLI view preserves the same repository, selected job, artifact, outcome, and limit.
PlannedPlanned local interface — illustrative.No customer data and no live product execution.
O
System Memorylocalhost / local only
Repository selected
Selected job
System Blueprint
Last run: not runOutput: outputs/blueprint.jsonCLI: prototype only
outputs/blueprint.jsonSystem identity and relationship graph
observed identities
24
relationships
31
runtime-only surfaces
2
↳ You no longer redraw the delivered architecture for the handover deck.
Action status: Planned local interface — illustrative.
Limit: The Blueprint cannot infer undocumented intent, unseen runtime behavior, or dynamic references beyond the declared observation boundary.
Selected job
Decision Memory
Last run: not runOutput: outputs/decisions.jsonInteraction: illustrative
outputs/decisions.jsonEvidence-linked decision record
decision
two approvers above threshold
evidence
PR-184 · ACME-91
business intent
owner confirmation required
↳ You no longer reconstruct the decision from memory or old chat threads months later.
Action status: Planned workflow — illustrative.
Limit: When recoverable evidence does not explain the rationale, Decision Memory marks it unknown and asks a human to confirm it.
Selected job
Durable Knowledge Links
Last run: not runOutput: outputs/bindings.jsonCLI foundation: verified
outputs/bindings.jsonBinding index with provenance trace
referent
prompt/refund-triage
bound knowledge
docs/runbook.md#refunds
dynamic references
review required
↳ You stop repairing guidance merely because a file, screen, or symbol moved.
Action status: Planned local interface — illustrative.
Limit: Ambiguous, generated, and runtime-only references can require review; matching an identity does not prove semantic equivalence.
Selected job
Behavior Baseline
Last run: not runOutput: outputs/baseline.jsonCLI foundation: verified
outputs/baseline.jsonVersioned probe baseline
probe policy
safe path validated
snapshot
representative output
production health
not observed
↳ You leave a repeatable reference for detecting behavior change instead of relying on memory.
Action status: Planned workflow — illustrative.
Limit: A baseline is not production monitoring or proof of correctness; it covers only approved probes and cannot trigger real side effects.
Selected job
Private Project Assistant
Last run: not runOutput: outputs/answers.jsonCLI foundation: verified
outputs/answers.jsonAnswer with provenance and knowledge state
answer state
known with citations
sources
PR-184 · routing.toml
unapproved authority
withheld
↳ Routine project questions start with inspectable evidence instead of returning immediately to you.
Action status: Planned local assistant — illustrative.
Limit: The assistant is not a generic chatbot or support desk; it refuses unsupported, stale, or out-of-bound questions.
Selected job
Handover Generator
Last run: not runOutput: outputs/handover/index.mdCLI foundation: verified
outputs/handover/index.mdDated handover and runbook package
sections
map · decisions · owners · gaps
verification
18 current · 1 review
approval
owner review required
↳ You review a generated handover instead of authoring a deck and runbook from scratch.
Action status: Planned workflow — illustrative.
Limit: The generated package is a reviewable draft; it cannot fill unknowns or transfer operational accountability by itself.
Selected job
Knowledge Coverage Check
Last run: not runOutput: outputs/gaps.jsonCLI foundation: verified
outputs/gaps.jsonCoverage and gap table
observed
24 identities
current coverage
19 identities
uncovered
3 targeted gaps
↳ You spend scarce subject-matter-expert time correcting missing areas instead of reviewing everything.
Action status: Planned local interface — illustrative.
Limit: Completeness is relative to the observed inventory and declared boundary, never a claim that the whole real-world system is documented.
Selected job
Coverage and Limitations Statement
Last run: not runOutput: outputs/coverage-boundary.mdCLI foundation: verified
outputs/coverage-boundary.mdCoverage and limitations statement
observed
repository + approved evidence
outside boundary
2 surfaces
attestation
not provided
↳ You can state the evidence boundary clearly instead of letting confidence imply coverage.
Action status: Planned local interface — illustrative.
Limit: The statement reflects declared access and current observations; it is not a certification, compliance attestation, or proof about unseen systems.
You spend scarce subject-matter-expert time correcting missing areas instead of reviewing everything.
Limit: Completeness is relative to the observed inventory and declared boundary, never a claim that the whole real-world system is documented.
Selected jobCoverage and Limitations StatementVerified foundation
adopt boundary --answers adopt/boundary-answers.json --archetype ai --write-statement ./onboard-output/coverage-boundary.md
Artifact
outputs/coverage-boundary.md
Outcome
You can state the evidence boundary clearly instead of letting confidence imply coverage.
Limit: The statement reflects declared access and current observations; it is not a certification, compliance attestation, or proof about unseen systems.
Give this to your coding agent
Inspect this repository using the public adopt CLI. Start with `adopt detect ./acme-support-agent --json`. Report what the command verifies, what remains unknown, and every local artifact created. Do not send repository content over the network.
adopt ui is not a public command. The browser Workbench shown here is a planned experience. Verified commands are labelled separately.
VERIFIED LOWER-LEVEL PRIMITIVES
Do not confuse foundation commands with finished experiences.
These commands were verified in public main as foundations for planned workflows. Their adjacent notes keep that distinction explicit.
adopt boundary --answers adopt/boundary-answers.json --archetype ai --write-statement ./onboard-output/coverage-boundary.md
Available boundary primitive; the guided Workbench review is planned.
CODING-AGENT PATH
Give the agent the evidence boundary too.
A useful instruction asks the agent to separate verified output from inference and announce writes before they happen.
Inspect ./acme-support-agent using only the public adopt CLI. Begin with `adopt detect ./acme-support-agent --json`. Before any command that may write, state the target path. Separate verified output from inference, list every unknown, and do not transmit repository content.