Agent-ready workflow · recon

Surface Map to Ranked Hypotheses

Turn authorized asset discovery into a deduplicated, technology-driven lead queue with one decisive next test per lead.

reconfingerprintingprioritizationlead-ledger
Stable IDworkflow:surface-to-hypothesisSource record dateHuman reviewNot recordedProvenanceCurated workflow graph

Enter with artifacts, leave with evidence.

Use when
  • new program
  • new asset
  • changed host
  • unknown stack
Prerequisites
  • dated scope snapshot
  • rate and automation rules
  • approved target roots
MCP retrieval

Agents can search this workflow, retrieve the complete graph, or request one stage through the read-only Vulns.co MCP connector.

01

Lock scope and provenance

Normalize included assets, exclusions, wildcard rules, and source dates before discovery.

Inputs
  • live program brief
  • approved roots
Outputs
  • scope snapshot
  • ambiguity queue
Evidence gate
Every candidate asset retains its scope source and retrieval date.
Negative control
An explicitly excluded asset is rejected by the same normalization rules.
Stop condition
Stop on unresolved ownership or scope ambiguity.
02

Fingerprint reachable stacks

Resolve live services and attach technology, version, title, certificate, and response evidence.

Inputs
  • approved roots
Outputs
  • live host inventory
  • stack evidence
Evidence gate
A technology label has a raw response, header, certificate, or service fingerprint behind it.
Negative control
A generic CDN or shared edge is not attributed to the application stack without origin evidence.
Stop condition
Stop active probing at the program rate limit.
03

Map operations and inputs

Collect reachable routes, parameters, APIs, JavaScript, identity states, and file or URL consumers.

Inputs
  • live hosts
  • controlled accounts
Outputs
  • endpoint matrix
  • input inventory
  • trust-boundary map
Evidence gate
Every operation records method, auth state, source artifact, and last observation.
Negative control
Dead, third-party, and client-only routes are labeled instead of treated as live server behavior.
Stop condition
Do not replay state-changing operations during mapping.
04

Rank and assign leads

Score leads by reachability, evidence, boundary importance, safe test cost, and chain prerequisites.

Inputs
  • stack evidence
  • endpoint matrix
  • trust map
Outputs
  • ranked lead ledger
  • one next discriminator per lead
Evidence gate
Each lead states one observation, one falsifiable hypothesis, one control, and one owner.
Negative control
Speculative leads without a reachable source or sink remain unassigned backlog.
Stop condition
Stop duplicate collection when an owner already has a decisive test.

Continue with the right depth.

References