Evidence the run may read
- Sanitized message headers and body
- Link and attachment analysis
- Recipient and report metadata
production blueprint · security
Phishing investigation packet agent blueprint.
Analyze sanitized message, sender, link, and user-report evidence into a bounded disposition and response checklist.
Search intent: AI agent for phishing investigation
Published by a2a cloudProduct-source methodology →
01 · the contract
Start from a bounded job, not a blank chat box.
trigger
A user or security control submits a suspected phishing event.
owner
security owner
agent stops at
A security responder approves blocking, takedown, or account containment actions.
KPI · decrease
phishing disposition time (minutes). Measure report submission to reviewed disposition.
outputs
Artifacts the run must produce
- Evidence timeline
- Maliciousness assessment
- Containment proposal
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.
- 01
intake
Validate and normalize the phishing investigation packet inputs.
- 02
message analyst
Produce the evidence timeline.
- 03
indicator correlator
Produce the maliciousness assessment.
- 04
incident reviewer
Challenge the phishing investigation packet result and prepare an approval packet.
- 05
security owner
A security responder approves blocking, takedown, or account containment actions.
intake→ validated input packet →message-analyst
message-analyst→ Evidence timeline →indicator-correlator
indicator-correlator→ Maliciousness assessment →incident-reviewer
incident-reviewer→ approval packet with evidence references →human-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.
case inputs
workspace/security/phishing-investigation-packet/inputs/**Read only the evidence attached to this workflow instance.
case outputs
workspace/security/phishing-investigation-packet/outputs/**Write drafts and evidence artifacts without modifying source records.
approved tools
security:phishing-investigation-packet:read-or-proposeInvoke only tools explicitly granted for this run; external writes remain gated.
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: security-phishing-investigation-packet
version: 0.1.0
entrypoint: agent:BlueprintAgent
expose:
public: false
composition:
planning: deterministic_dag
max_nodes: 6
max_parallel: 1
max_replans: 005 · acceptance test
Pass only with evidence.
- 01
Evidence traceability
Every material conclusion cites an input artifact or a scoped tool result from this run.
- 02
The disposition distinguishes verified indicators, reputation signals, and analyst inference.
The disposition distinguishes verified indicators, reputation signals, and analyst inference.
- 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.
Attachments or headers cannot be safely inspected within the granted evidence set.
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": "security-phishing-investigation-packet",
"caller": "user:workflow-owner",
"task_id": "case_phishing_investigation_packet",
"skill_name": "phishing_investigation_packet",
"input_hash": "4d7c...9a2f",
"grant_ids": [
"grt_case_inputs",
"grt_tool_propose"
],
"status": "ok",
"result_preview": "Output prepared: Evidence timeline. Human decision remains separate.",
"elapsed_ms": 4218
}related blueprints
Keep the operating model consistent.
Security
Vulnerability triage
Combine scanner evidence, asset context, exploitability signals, and compensating controls into a reviewed priority decision.
Security
Secrets exposure response
Build an exposure timeline, affected-resource list, and rotation plan without copying secret values into artifacts.
Security
Access review evidence
Normalize account and entitlement exports, flag anomalous access, and produce a reviewer-ready certification packet.
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.