| Agent | Purpose | Key Constraint |

|----------------------|-------------------------------------------------------------------|------------------------------------------------|

| **rpi-agent** | Autonomous agent with subagent delegation for complex tasks | Requires `runSubagent` tool enabled |

| **task-researcher** | Produces research documents with evidence-based recommendations | Research-only; never plans or implements |

| **task-planner** | Creates 3-file plan sets (plan, details, prompt) | Requires research first; never implements code |

| **task-implementor** | Executes implementation plans with subagent delegation | Requires completed plan files |

| **task-reviewer** | Validates implementation against research and plan specifications | Requires research/plan artifacts |

| Agent | Purpose | Key Constraint |

|---------------------------|--------------------------------------------------------------------|-----------------------------------------------|

| **prd-builder** | Creates Product Requirements Documents through guided Q&A | Iterative questioning; state-tracked sessions |

| **brd-builder** | Creates Business Requirements Documents with reference integration | Solution-agnostic requirements focus |

| **adr-creation** | Interactive ADR coaching with guided discovery | Socratic coaching approach |

| **security-plan-creator** | Creates comprehensive cloud security plans from blueprints | Blueprint-driven threat modeling |

| **doc-ops** | Documentation operations and maintenance | Does not modify source code |

| Agent | Purpose | Key Constraint |

|------------|--------------------------------------------|---------------------------------------|

| **memory** | Persists repository facts for future tasks | Stores only durable, actionable facts |

| Agent | Purpose | Key Constraint |

|--------------------|--------------------------------------------------|---------------------------------------|

| **pr-review** | 4-phase PR review with tracking artifacts | Review-only; never modifies code |

| Agent | Purpose | Key Constraint |

|-----------------------------|----------------------------------------------------|--------------------------------------|

| **gen-jupyter-notebook** | Creates structured EDA notebooks from data sources | Requires data dictionaries |

| **gen-streamlit-dashboard** | Develops multi-page Streamlit dashboards | Uses Context7 for documentation |

| **gen-data-spec** | Generates data dictionaries and profiles | Produces JSON and markdown artifacts |

| **arch-diagram-builder** | Builds ASCII block diagrams from Azure IaC | Parses Terraform, Bicep, ARM scripts |

| Agent | Purpose | Key Constraint |

|--------------------------|------------------------------------------------------------|-------------------------------------------|

| **github-issue-manager** | Interactive GitHub issue filing and navigation | Uses MCP GitHub tools |

| **ado-prd-to-wit** | Analyzes PRDs and plans Azure DevOps work item hierarchies | Planning-only; does not create work items |

| **hve-core-installer** | Decision-driven HVE-Core installation with 6 methods | Environment detection and validation |

| Agent | Purpose | Key Constraint |

|------------------------------|---------------------------------------------|----------------------------------------|

* `.copilot-tracking/research/{{YYYY-MM-DD}}-topic-research.md` (primary research with evidence-based recommendations)

* `.copilot-tracking/subagent/{{YYYY-MM-DD}}/task-research.md` (subagent research outputs when delegating)

* `.copilot-tracking/plans/{{YYYY-MM-DD}}-task-plan.instructions.md` (implementation plan with checklist items)

* `.copilot-tracking/details/{{YYYY-MM-DD}}-task-details.md` (step-by-step execution details)

* `.copilot-tracking/changes/{{YYYY-MM-DD}}-task-changes.md` (chronological log with Added/Modified/Removed sections)

* `.copilot-tracking/reviews/{{YYYY-MM-DD}}-{{topic}}-review.md` (findings with severity levels and follow-up work)

* `.github/instructions/*.instructions.md` (coding guidelines and conventions)

* `.github/prompts/*.prompt.md` (reusable workflow prompts)

* `.copilot-tracking/sandbox/{{YYYY-MM-DD}}-{{prompt-name}}-{{run-number}}/execution-log.md` (test execution trace)

* `.copilot-tracking/sandbox/{{YYYY-MM-DD}}-{{prompt-name}}-{{run-number}}/evaluation-log.md` (quality validation results)

* `.copilot-tracking/pr/review/{normalized-branch}/in-progress-review.md` (living review document with findings)

* `.copilot-tracking/pr/review/{normalized-branch}/pr-reference.xml` (PR metadata and diff summary)

* `.copilot-tracking/pr/review/{normalized-branch}/handoff.md` (finalized comments for PR submission)

* `docs/prds/<kebab-case-name>.md` (PRD document with requirements)

* `.copilot-tracking/prd-sessions/<kebab-case-name>.state.json` (session state for resume capability)

* `docs/brds/<kebab-case-name>-brd.md` (BRD document with business objectives)

* `.copilot-tracking/brd-sessions/<kebab-case-name>.state.json` (session state for resume capability)

**Creates:** Documentation updates and maintenance artifacts:

* `.copilot-tracking/doc-ops/{{YYYY-MM-DD}}-session.md` (session tracking for documentation operations)

**Creates:** Repository memory records and session context:

* `.copilot-tracking/memory/{{YYYY-MM-DD}}/{{short-description}}-memory.md` (session continuity context)

* `.copilot-tracking/memory/{{YYYY-MM-DD}}/{{short-description}}-artifacts/` (optional companion files)

* `/memories/repo/<descriptive-name>.jsonl` (durable repository facts for future tasks)

**Workflow:** Identify actionable repository fact → Validate durability → Store with context → Available for future tasks

**Critical:** Stores only durable, reusable facts. Does not store transient discussion, personal preferences, or speculative information.

* `.copilot-tracking/plans/security-plan-{blueprint-name}.plan.md` (planning artifacts and threat analysis)

* `security-plan-outputs/security-plan-{blueprint-name}.md` (final security plan document)

* `notebooks/*.ipynb` (EDA notebooks with parameterized data loading)

* `data/processed/*.parquet` (derived datasets with semantic naming)

* `app.py` (main entry point with page navigation)

* `pages/*.py` (summary statistics, univariate/multivariate analysis, time series)

* `requirements.txt` (pinned dependencies)

* `outputs/data-dictionary-{{dataset}}-{{YYYY-MM-DD}}.md` (column definitions and semantics)

* `outputs/data-profile-{{dataset}}-{{YYYY-MM-DD}}.json` (statistical profile for downstream tools)

* `outputs/data-objectives-{{dataset}}-{{YYYY-MM-DD}}.json` (analysis goals and constraints)

* `outputs/data-summary-{{dataset}}-{{YYYY-MM-DD}}.md` (human-readable overview)

* Inline ASCII block diagrams embedded in markdown (pure ASCII for consistent alignment)

* Component legend and relationship key

* `.copilot-tracking/github-issues/issue-{number}.md` (issue creation logs with API responses)

* `.copilot-tracking/github-issues/issues-list-{timestamp}.md` (navigation session results)

* `.copilot-tracking/github-issues/search-{timestamp}.md` (search query results)

* `.copilot-tracking/github-issues/session-state.md` (active context and filter state)

* `.copilot-tracking/github-issues/draft-issue.md` (working draft before submission)

* `.copilot-tracking/github-issues/current-filters.md` (saved filter configurations)

* `.copilot-tracking/workitems/prds/<artifact-normalized-name>/planning-log.md` (session activity and decisions)

* `.copilot-tracking/workitems/prds/<artifact-normalized-name>/artifact-analysis.md` (PRD parsing and extraction)

* `.copilot-tracking/workitems/prds/<artifact-normalized-name>/work-items.md` (Epic/Feature/Story hierarchy)

* `.copilot-tracking/workitems/prds/<artifact-normalized-name>/handoff.md` (final handoff for ADO creation)

* Settings updates in `.vscode/settings.json` (VS Code configuration)

* `.vscode/mcp.json` (MCP server configuration when applicable)

* Test results summary (pass/fail counts by category)

* Issue registry with reproduction steps (severity-categorized findings)

* Performance metrics (page load times, render benchmarks)