Agentic system analysis overview

Type: types/type-spec.md

The reading entry point of one analyse-agentic-system run's retained set. It holds identity, evidence boundary, source register, an amendment index, synthesis, limitations and both verifications. Runtime, memory, epistemic and reconciliation findings live in their respective members. The sibling ARTIFACT.yaml selects the analysis set type and pins every member, including this overview.

Frontmatter

Field	Required	Use
type	Yes	agentic-systems/types/agentic-system-analysis-overview.md
description	Yes	For a complete run, the synthesizer's one-sentence retrieval description of the system's mechanism and limits, which the public review also carries; otherwise a code-written description naming the system, selected boundary, and disposition
run-id	Yes	Canonical AAS-YYYY-MM-DD-system-slug-nn identity allocated by the producing skill
system	Yes	Source-native system name or the caller's unambiguous identifier
run-date	Yes	Date the run opened
result-disposition	Yes	complete, blocked, or out-of-scope
target-class	Yes	Target role under the boundary contract; null before classification
boundary-kind	Yes	Extent of the selected target under the boundary contract; null before establishment
reviewed-boundary	Yes	Immutable revision or capture identity shared by the run, or null before one could be established
analysis-cutoff	Yes	Applicability cutoff for the frozen evidence, or null before one could be established
evidence-tier	Yes	code-grounded, doc-grounded, or null before the runtime baseline could support a tier
inputs-commit	Yes	Full repository commit supplying the instructions, contracts, schemas and package code; publication requires the method unchanged

For a complete run, the five boundary fields are non-null. A blocked or out-of-scope overview retains every required section and states what was not reached, why, and which conclusion that prevents.

Shared contracts

The boundary contract defines boundary classifications and source declarations. The source contract defines evidence layers, anchors and quotations. The record contract defines the namespace, declarations, annotations, amendments, fields and status meanings for all members.

Identity and completion

Working output lives under kb/agentic-systems/reports/state/<run-id>/output/; publication retains it under kb/agentic-systems/reports/retained/<run-id>/. Run state and compact reviews pin ARTIFACT.yaml. The manifest pins the reports, and inputs-commit identifies the method's committed inputs. The set type owns membership and set checks. File validation checks this overview independently; directory validation checks the whole set. Only complete sets may publish or supply comparison rows. Correct retained output through a new run.

Required sections

Boundary and evidence

## Boundary and evidence contains the boundary account defined by the boundary contract, copied from the boundary job.

Source register

## Source register contains that job's SRC-* rows under the boundary contract. Source identity and evidence scopes remain stable across members. For a complete run, code appends Amended or superseded records: <IDs or none> and a link to reconciliation.md. Resolve these IDs through that member before relying on an analyst's original wording.

Bounded synthesis

## Bounded synthesis gives the evidence basis and boundary, architectural characterization and claimed work, runtime map, only the discriminating mechanisms this target needs, scenario-relative assessment, and concrete evidence or system changes that would alter the assessment. Where the runtime report supports it, the synthesis states separately whether the system meets theory-builder conditions 1–4, whether criticism of a consumed theory improved the system's capacity for future action, whether the system is reflective or autonomous, and whether it is self-improving at the declared boundary, each at its own evidence status; these are independent properties, not a grade or a ladder. It is organized around the system's operational progression, not as concatenated analyst reports, and cites member records rather than restating them. It gives no product ranking, generic adoption advice, system-wide epistemic grade, Commonplace delta, or transfer recommendation. For learning and self-improvement findings it leads with the strongest supported contribution, including partial results, then states the unresolved question, at the level of the comparison actually performed.

The Bounded synthesis has a second reader. Code publishes it, after an evidence-basis line and before the Limitations, as the body of the public generated review. It must therefore read without the members' context: a public reader has the IDs it cites and its links, which code rewrites to point into the retained set, but not the members' prose.

Limitations

## Limitations contains one row per limitation:

limitation | affected source, record, or route IDs | inspected boundary | conclusion prevented | evidence that would resolve it

Carry every Unresolved conflict: in the reconciliation into this section. A record fault found during synthesis verification is stated here with its prevented conclusion; it does not reopen reconciliation. If representing it as a limitation would make the synthesis misleading, stop the run. Use none only after checking the whole set. A blocker is also represented under Verification and blockers; this section still states its analytical consequence.

Verification and blockers

## Verification and blockers contains ### Record verification, ### Synthesis verification, ### Deterministic validation, and ### Blockers. The two independent checks remain separate: record verification covers the memory profile, amendments, supersessions and scope; synthesis verification covers support for public statements, readability without the members' context, and unresolved conflicts in Limitations. Record the exact deterministic validation targets and results for every member, and every unresolved blocker. Do not record projection review jobs, publication attempts, or cleanup here. A complete run says none under blockers; a blocked or out-of-scope run states its stopping condition here. The overview's digest never appears inside its own bytes.

Template

---
type: agentic-systems/types/agentic-system-analysis-overview.md
description: "{one sentence on the system's mechanism and limits}"
run-id: AAS-YYYY-MM-DD-system-slug-nn
system: "{source-native system name}"
run-date: "YYYY-MM-DD"
result-disposition: complete
target-class: "{selected target class}"
boundary-kind: whole-system
reviewed-boundary: "{immutable revision or capture identity}"
analysis-cutoff: "YYYY-MM-DD"
evidence-tier: code-grounded
inputs-commit: "{full commit of this repository at run start}"
---

# {System} agentic-system analysis

## Boundary and evidence

## Source register

## Bounded synthesis

## Limitations

## Verification and blockers

### Record verification

### Synthesis verification

### Deterministic validation

### Blockers