Skip to content
DevSquad Copilot
This project is under active development and subject to breaking changes. See the changelog for release notes.

Glossary

A

Section titled “A”

ADR (Architecture Decision Record) A document that records a significant technical decision, including context, priorities, options evaluated, and the chosen approach with trade-offs. ADRs follow the lifecycle: Proposed, then Accepted, then Superseded by NNNN. See Decisions.

AI Cost Posture A gated section in the feature spec template, required when the spec declares Describes AI capability: yes. Captures five operational commitments: model-tier (Reasoning, Frontier, Mid, or Fast per AI step), latency budget (p50, p95, p99 plus behavior on breach), prompt-stability invariant, per-call cost ceiling, and cost-incident escalation. For platforms where the runtime picks the model, fields are set to N/A with a one-line note. See AI Capability Specs.

C

Section titled “C”

Behavioral Constraints (agent body section) A ## Behavioral Constraints section on a user-facing agent file listing rules the agent must honor even when its runtime tools: array would permit otherwise (for example, “never APPROVE on a PR”; “never commits to integration branch”). The tools: list is the runtime authority boundary; this section captures only what the runtime cannot enforce. Present on the seven user-facing agents (devsquad, devsquad.implement, devsquad.plan, devsquad.review, devsquad.refine, devsquad.specify, devsquad.decompose); worker sub-agents inherit their envelope from the parent.

Comprehension Checkpoint A verification step triggered before medium or high impact tasks. The agent asks the developer to describe what will happen in their own words, ensuring active understanding rather than passive delegation. Generic responses (“ok”, “go”) trigger deeper questions. See Comprehension Checkpoints.

Conductor The central devsquad agent that detects user intent and delegates to specialist agents. It contains no domain logic, only routing and orchestration. See Conductor.

Conformance Criteria Testable conditions (identified by CC-XXX format) that verify a specification requirement is met. Each includes a scenario, input, and expected output.

Coordinator Agent A specialist agent that delegates internally to focused worker sub-agents with isolated context. The framework has four coordinators: plan, implement, review, and refine.

Composition (agent body section) A ## Composition section on a coordinator agent file declaring load-bearing cross-component invariants between the coordinator and its typed sub-agents (for example, validate runs before execute for Medium and High impact tasks). The runtime surfaces each sub-agent’s description: at invocation time, so the section does not re-list sub-agents.

D

Section titled “D”

Disk Artifacts Versioned files on disk that serve as the source of truth for the framework. Includes spec.md, plan.md, tasks.md, and ADRs. Session context is ephemeral; artifacts persist via git.

H

Section titled “H”

Handoff Envelope A structured context-passing contract between agent phases (ADR 0003). Contains four sections: relevant artifacts, inherited assumptions, pending decisions, and discarded information. See Reasoning and Handoff.

I

Section titled “I”

IDD (Intent-Driven Development) The methodology that structures vision, specification, and architecture decisions before code generation. Built on three pillars: Intent (Why), Specification (What), and Architecture Decisions (How). IDD is the overarching methodology. SDD (Spec-Driven Development) describes the execution model within it, where specifications drive all downstream work.

Impact Classification A three-level system for classifying changes before acting. Low (typo, log) means direct action. Medium (function change) means plan + confirm. High (service, API, schema) means ADR + explicit approval. See Impact Classification.

M

Section titled “M”

MCP (Model Context Protocol) A protocol for providing tools to AI agents. MCP servers expose capabilities (GitHub, Azure DevOps, Azure, Microsoft Learn, Draw.io) that agents call to interact with external systems. See MCP Servers.

O

Section titled “O”

OWASP Open Worldwide Application Security Project. The framework references OWASP guidelines for code-level security assessment.

Q

Section titled “Q”

Quality Gate An automated validation checkpoint that verifies artifact completeness and consistency before delivery. Applied to specs, ADRs, task decompositions, and code implementations.

R

Section titled “R”

Reasoning Log A record of decisions made during agent execution. Each entry includes: the decision, principles applied, alternatives considered, justification, and confidence level (High/Medium/Low). Used for auditability and handoff between phases. See Reasoning and Handoff.

S

Section titled “S”

SDD (Spec-Driven Development) The development approach where specifications drive all downstream work: planning, decomposition, implementation, and review. Specifications define WHAT and WHY; implementation artifacts define HOW.

Socratic AI An adaptive approach where agents ask clarifying questions, surface trade-offs, and present options rather than making autonomous decisions. Low-impact tasks are fast-tracked; high-impact tasks require comprehension verification. Based on ADR 0005.

Spec Evolution Log A required section in every feature and migration spec that records amendment history. Each row carries version, date, change summary, trigger (new work, drift, external constraint, one of three failure (<category>) values, or other (<short reason>)), and author. The log is the durable signal that the spec changed and why. See Spec Amendment.

Specialist Agent An agent that owns the logic for a specific delivery phase (e.g., devsquad.specify for specifications, devsquad.implement for code). Can be invoked directly or through the conductor.

STRIDE A threat modeling methodology (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, Elevation of Privilege) used by the security review process.

T

Section titled “T”

Tool Extension A mechanism for injecting MCP server tools into existing plugin agents via YAML patches (Preview, ADR 0010). Consumer creates .github/devsquad/tool-extensions/*.yml, and a sync script generates workspace-level agent overrides.

U

Section titled “U”

Upstream Fix (Failure Triage) The discipline of routing an agent-originated failure to the upstream artifact that owns the broken rule, instead of patching the prompt. Three categories: failure (spec) (fix in spec.md, ADR, or glossary), failure (validation) (fix in conformance criteria, tests, or rubric), failure (agent) (fix in agent body, composition declaration, or tool config). Prompt patches that mask a symptom without reconciling the upstream artifact are forbidden. See Spec Amendment.

W

Section titled “W”

Worker Sub-agent A specialized agent invoked by a coordinator agent to perform a focused subtask in an isolated context window. Workers return results to the coordinator and do not interact with the user directly.

Section titled “What to Read Next”