Proposed specificationSource-located exemplarH1

Prompt chaining

Pass intermediate results through planned stages.

JOIN POINTsequential_stage_compositionRECORD STATUSCandidate · v0.1.0

UML class view

The proposed contract drawn as participant roles, received calls, and numbered dependencies.

Object interactions

The same roles and calls as a class collaboration and a numbered message sequence.

Object interaction diagram for Prompt chaining, showing role collaboration and a numbered message sequence — FIG. W01 / OBJECT INTERACTIONS
Open full SVG Download PDF

Pointcut

Where the mechanism applies. All four dimensions belong to the proposed match, not just its vocabulary.

Context: A task is decomposed into ordered agent or prompt stages.
Trigger: One stage produces an intermediate result consumed by the next.
Structure: Stage graph, typed handoff payload, validation, and terminal output.
Effect: Stages execute in order and downstream input is attributable to upstream output.

Protected contract

The obligation that clients, operators, and auditors should be able to rely on.

Requires: Stage order and handoff schema are defined.
Ensures: Every stage receives validated input and the final output identifies its stage lineage.
Invariant: A later stage cannot execute before its prerequisites complete.
Violation: A stage failure stops, retries, or branches according to declared policy.

Negative control

Independent prompts with no intermediate-result dependency do not count.

Variation & lifecycle

The hot spot is admitted only where the protected semantics survive.

Permitted variation

Sequential agents, prompt templates, pipelines, and transform nodes.

Test obligation

Test order, input propagation, stage failure, and branching.

Evolution question

Did chains add typed handoffs and explicit failure propagation?

Prompt + code design

An illustrative prompt and a small design sketch for the proposed contract. Names and APIs are invented; this is not a repository excerpt.

PROMPT SKETCH

Run the named stages in order. Validate each handoff against its declared schema, and make the final output identify the stage lineage that produced it.

CODE DESIGN / PYTHON

def run_chain(stages, initial_input):
    payload, lineage = initial_input, []
    for stage in stages:
        validated = stage.input_schema.validate(payload)
        payload = stage.run(validated)
        lineage.append(stage.id)
    return ChainResult(output=payload, lineage=lineage)

Illustrative design sketch. A chain without validated handoffs is a sequence of unrelated prompts. This example is a teaching aid, not implementation evidence or a conformance claim.

Detection & matching

The frozen rule that retrieves this record, the canonical source exemplar, and the control that should stay unmatched.

Join point: sequential_stage_composition
Source locator: Anthropic
Lexical cue: (?i)\b(?:prompt[_-]?chain(?:ing)?|chain[_-]?(?:prompt|agent)|sequential[_-]?(?:workflow|pipeline))\bFrozen expression, v0.1.0. A match is a retrieval cue, not a confirmed implementation.
Aliases: promptchain, sequentialagent, sequentialpipeline
Canonical exemplar: google/adk-pythonsrc/google/adk/agents/sequential_agent.py:91Support: tests/unittests/agents/test_sequential_agent.py:80
Negative control: Independent prompts with no intermediate-result dependency do not count.

Scope-anchored review. pohe-agent analyze <repo> --pattern W01 anchors the cue to its enclosing declaration and checks a separate operational signature for all four pointcut dimensions. Missing roles and negative-control risks remain visible; a structural match is not E2/E3. Calibration report ↗

Repository files

Up to five public files retrieved by the frozen cue and pinned to the audited commit. A match is a place to inspect, not a confirmed implementation.

01google/adk-pythonsrc/google/adk/agents/sequential_agent.py:91CANONICAL SOURCE 02aaif-goose/goosedocumentation/docs/guides/sessions/in-session-actions.md:123RETRIEVAL CUE 03affaan-m/ECCskills/autonomous-loops/SKILL.md:32RETRIEVAL CUE 04agno-agi/agnolibs/agno/tests/unit/db/test_reserved_run_metadata.py:117RETRIEVAL CUE 05microsoft/agent-frameworkpython/packages/devui/tests/devui/conftest.py:460RETRIEVAL CUE

Evidence & provenance

Each layer answers a different question. Do not substitute retrieval for review.

Source-located exemplar

A concrete source anchor is available for this proposed record. It is one exemplar, not evidence that all repositories—or every contract field—conform.

Review boundary: this candidate and its specification come from an AI-assisted study. Independent human validation remains open.

ANY-EVIDENCE FAMILY COVERAGE6/36

16.7% of repository families in the frozen panel have at least one current cue, historical cue, or canonical source layer. This is not confirmed prevalence.

Current lexical-cue repositories: 7 / 41

Canonical source anchorgoogle/adk-python 946ca77224d5src/google/adk/agents/sequential_agent.py:91

Supporting surfacetests/unittests/agents/test_sequential_agent.py:80

SequentialAgent passes output through ordered sub-agents.

Exploratory retrieval cue Regex / v0.1.0

(?i)\b(?:prompt[_-]?chain(?:ing)?|chain[_-]?(?:prompt|agent)|sequential[_-]?(?:workflow|pipeline))\b

This expression is a discovery instrument only. A match is a retrieval cue—not a confirmed implementation, pointcut match, or contract proof.