a2a cloud
production blueprint · compliance & audit

Control evidence collection agent blueprint.

Map a control request to approved sources, collect a bounded evidence packet, and flag missing attestations.

Search intent: AI agent for compliance evidence collection

Published by a2a cloudProduct-source methodology →
01 · the contract

Start from a bounded job, not a blank chat box.

trigger

A scheduled control test or evidence request opens.

owner

compliance owner

agent stops at

The control owner attests completeness and approves release to testing or audit.

KPI · increase

evidence requests accepted first pass (percent). Measure packets accepted without a completeness follow-up.

inputs

Evidence the run may read

  • Control and period definition
  • Approved evidence source index
  • Owner and retention guidance
outputs

Artifacts the run must produce

  • Indexed evidence packet
  • Control-to-evidence map
  • Missing evidence and owner list
02 · topology

Small specialists. Named handoffs. One accountable decision.

Each stage produces an artifact another stage can inspect. The final node is a person, not an autonomous write to an external system.

  1. 01

    intake

    Validate and normalize the control evidence collection inputs.

  2. 02

    request interpreter

    Produce the indexed evidence packet.

  3. 03

    evidence collector

    Produce the control-to-evidence map.

  4. 04

    control owner reviewer

    Challenge the control evidence collection result and prepare an approval packet.

  5. 05

    compliance owner

    The control owner attests completeness and approves release to testing or audit.

intakevalidated input packetrequest-interpreter
request-interpreterIndexed evidence packetevidence-collector
evidence-collectorControl-to-evidence mapcontrol-owner-reviewer
control-owner-reviewerapproval packet with evidence referenceshuman-approver
03 · authority

Grant the run only what this case needs.

Source material is read-only. Drafts land in a case-specific output path. Tools may read or propose; the human gate owns the external write.

read

case inputs

workspace/compliance/control-evidence-collection/inputs/**

Read only the evidence attached to this workflow instance.

write-output

case outputs

workspace/compliance/control-evidence-collection/outputs/**

Write drafts and evidence artifacts without modifying source records.

invoke-scoped-tool

approved tools

compliance:control-evidence-collection:read-or-propose

Invoke only tools explicitly granted for this run; external writes remain gated.

required human decision

The control owner attests completeness and approves release to testing or audit.

Decision owner: compliance owner.

04 · implementation

A private, bounded starting manifest.

The blueprint starts private, caps its DAG, disables replanning, and exposes no public endpoint. Add only the tools and data adapters this workflow has approved.

a2a.yamlsafe starting point
name: compliance-control-evidence-collection
version: 0.1.0
entrypoint: agent:BlueprintAgent
expose:
  public: false
composition:
  planning: deterministic_dag
  max_nodes: 6
  max_parallel: 3
  max_replans: 0
05 · acceptance test

Pass only with evidence.

  1. 01

    Evidence traceability

    Every material conclusion cites an input artifact or a scoped tool result from this run.

  2. 02

    Every artifact has source, period, owner, retrieval date, and direct control relevance.

    Every artifact has source, period, owner, retrieval date, and direct control relevance.

  3. 03

    Approval boundary

    The run stops at a proposal and records the human decision before any external side effect.

failure containment

Stop small. Preserve the evidence.

Evidence is outside period, expired, inaccessible, or lacks accountable ownership.

Containment: Return a partial result with unresolved items; do not broaden scope or perform an external write.

Operator: Attach the missing evidence, narrow the brief, or explicitly approve a new scoped run.

A required input or tool grant is unavailable.

Containment: Stop the affected branch and preserve completed artifacts in the case output workspace.

Operator: Grant only the missing resource or continue with that branch marked out of scope.

06 · proof

Sign the run facts. Keep money in the billing ledger.

Current platform receipts sign caller identity or classification, skill, bounded input evidence, verified grant IDs when present, outcome or result preview, and timing. Optional file, tool, artifact, handoff, evaluation, and review fields require separate instrumentation and are not populated by default. Price, fees, payouts, and later human approvals remain separate platform records.

The example uses only fields populated by the current platform sealing paths. It is illustrative, not a record of a real customer run.

ExecutionReceipt · selected fieldsEd25519 token
{
  "receipt_id": "rcpt_01J...",
  "schema_version": 1,
  "agent_name": "compliance-control-evidence-collection",
  "caller": "user:workflow-owner",
  "task_id": "case_control_evidence_collection",
  "skill_name": "control_evidence_collection",
  "input_hash": "4d7c...9a2f",
  "grant_ids": [
    "grt_case_inputs",
    "grt_tool_propose"
  ],
  "status": "ok",
  "result_preview": "Output prepared: Indexed evidence packet. Human decision remains separate.",
  "elapsed_ms": 4218
}
put the pattern to work

Start private. Scope the authority. Require the decision.

Deploy the workflow as a bounded internal agent, verify its outputs and the receipt fields actually emitted, then expand only the scopes your acceptance test proves it needs.