Skip to content
Back to skills

Project Context Setup

ASecurity

Scaffolds per-repository agent context so coding agents share the same issue tracker rules, triage label vocabulary, domain glossary, ADR layout, and handoff conventions. Triggers on: "set up project context", "configure agent docs", "create CONTEXT.md", "setup agent workflow", "agent issue tracker setup", "triage labels", "domain glossary for agents". Use when a repo needs durable context files before planning, triage, debugging, TDD, architecture review, or multi-agent implementation.

  • 318 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added May 27, 2026
developmentgodebugginggit

Works with

  • cli

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned May 27, 2026

npx -y skills add Mathews-Tom/armory --skill project-context-setup --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Project Context Setup?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Project Context Setup
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/mathews-tom-project-context-setup/badge)](https://www.skillsdirectory.com/skills/mathews-tom-project-context-setup)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: project-context-setup
description: 'Scaffolds per-repository agent context so coding agents share the same issue tracker rules, triage label vocabulary, domain glossary, ADR layout, and handoff conventions. Triggers on: "set up project context", "configure agent docs", "create CONTEXT.md", "setup agent workflow", "agent issue tracker setup", "triage labels", "domain glossary for agents". Use when a repo needs durable context files before planning, triage, debugging, TDD, architecture review, or multi-agent implementation.'
metadata:
  version: 1.0.0
  category: development
  tags: [agent-context, project-setup, domain-glossary, triage]
  difficulty: intermediate
  phase: define
---

# Project Context Setup

Scaffold the repo-local context files that make agent workflows consistent across sessions:
issue tracker rules, triage labels, domain glossary layout, ADR layout, agent brief format,
and out-of-scope memory.

## When To Use

| Use this skill | Use another package |
| -------------- | ------------------- |
| A repo lacks `docs/agents/` context files | `adr-writer` for one architecture decision |
| Agents need consistent domain vocabulary | `task-decomposer` for an already-scoped feature |
| Triage or issue publishing needs label conventions | `github` for direct GitHub CLI operations |
| Multi-agent work needs durable issue briefs | `handoff` for session-local continuation notes |

## Reference Files

| File | Contents | Load When |
| ---- | -------- | --------- |
| `references/context-format.md` | `CONTEXT.md` and `CONTEXT-MAP.md` glossary format | Creating or editing domain docs |
| `references/agent-brief.md` | Durable `ready-for-agent` issue brief contract | Preparing issue handoff |
| `references/out-of-scope.md` | Persistent memory for rejected enhancements | Triage includes `wontfix` decisions |

## Core Workflow

### 1. Explore

Inspect the actual repo state before writing:

- `git remote -v` and `.git/config` for issue tracker hints
- root `AGENTS.md` and `CLAUDE.md`
- root `CONTEXT.md` and `CONTEXT-MAP.md`
- `docs/adr/` and context-scoped ADR directories
- existing `docs/agents/`
- existing `.out-of-scope/`

Report what exists and what is missing. Do not overwrite existing files blindly.

### 2. Choose Setup Values

Resolve these values with the user or from clear repo evidence:

| Decision | Default | Output |
| -------- | ------- | ------ |
| Issue tracker | GitHub if a GitHub remote exists, otherwise local markdown | `docs/agents/issue-tracker.md` |
| Triage labels | Canonical labels equal role names | `docs/agents/triage-labels.md` |
| Domain docs | Single root context | `docs/agents/domain.md` |
| Agent instructions file | Existing `AGENTS.md`, else existing `CLAUDE.md` | Updated `## Agent context` block |

If both `AGENTS.md` and `CLAUDE.md` exist, update `AGENTS.md` for cross-agent portability.
If neither exists, ask before creating one.

### 3. Write Context Files

Create or update these files:

```text
docs/agents/
├── issue-tracker.md
├── triage-labels.md
└── domain.md
```

Use these conventions:

- Keep issue tracker instructions operational: exact CLI, label mutation policy, and whether agents may comment or close issues.
- Map canonical triage roles to real labels. Do not create duplicate labels when the repo already has equivalents.
- Document whether the repo uses a single root `CONTEXT.md` or `CONTEXT-MAP.md` with per-context glossaries.
- Record ADR lookup rules. Use `docs/adr/` for global decisions and context-local `docs/adr/` directories where present.

### 4. Add Agent Context Block

Add or update this block in the chosen instructions file:

```markdown
## Agent context

### Issue tracker

See `docs/agents/issue-tracker.md` for where issues live and which operations agents may perform.

### Triage labels

See `docs/agents/triage-labels.md` for canonical triage roles and their concrete label strings.

### Domain docs

See `docs/agents/domain.md` for `CONTEXT.md`, `CONTEXT-MAP.md`, and ADR lookup rules.
```

Update the existing block in place if it already exists. Do not append duplicates.

### 5. Optional Domain Glossary

Create `CONTEXT.md` only when the user has resolved at least one domain term. Use
`references/context-format.md`. Do not fill it with generic programming concepts.

Create `CONTEXT-MAP.md` only for multi-context repos where separate glossaries are needed.

## Verification

- `docs/agents/issue-tracker.md` exists and states the tracker, allowed operations, and required tools.
- `docs/agents/triage-labels.md` maps `needs-triage`, `needs-info`, `ready-for-agent`, `ready-for-human`, and `wontfix`.
- `docs/agents/domain.md` states single-context or multi-context layout and ADR lookup rules.
- The chosen instructions file contains exactly one `## Agent context` block.
- Existing user-authored content outside that block is preserved.

Files in this skill

  • SKILL.md4.8 KB
  • evals/cases.yaml1.9 KB
  • references/agent-brief.md1.8 KB
  • references/context-format.md1.8 KB
  • references/out-of-scope.md1.3 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.