security-game/AGENTS.md

AGENTS.md

Purpose

This document defines the development and collaboration workflow for agents and human contributors working on the security guard game.

The project should be developed in small, reviewable steps. Agents must keep planning, task status, session memory, and project documentation up to date as work progresses.

Source-of-Truth Files

• PLAN.md — The general product and game-design plan. Update it when the intended scope, mechanics, or priorities change.
• TODO.md — The active task list. Organize work into sections and use checkbox items for every actionable task.
• DONE.md — Completed task sections moved from TODO.md. Preserve the original section structure and checkboxes when moving finished work.
• MEMORY.md — Persistent session memory for important decisions, discoveries, unresolved questions, constraints, and useful implementation context.
• README.md — Root-level documentation for the project. It should explain what the game is, how to work on it, how to run it, and where to find project information.
• AGENTS.md — This collaboration and agent workflow guide.
• CHANGELOG.md — The version history of user-facing changes. Keep an entry for each patch, minor, and major version.
• docs/development-status.md — The implemented-features list and what is not yet implemented. Update it whenever a feature lands, is renamed, or is removed.
• docs/architecture.md — The arc42-style architecture documentation (goals, constraints, building blocks, runtime views, decisions, risks).
• docs/gameplay.md — Gameplay systems, venues, and the escalation model.
• docs/asset-tooling.md — Asset pipelines, model tools, and screenshot gates.
• docs/npc-model-sourcing.md — Character-source evaluation, licence rules and the character-development plan.
• docs/getting-started.md — Setup, run/test commands, and the controls table.
• docs/smoke-checklist.md — The manual smoke-test checklist.
• docs/adr/ — Architecture Decision Records.

Priority and Versioning

TODO.md is the priority source. The active order is:

1. unversioned development tracks (currently 3D model re-evaluation, character development and local map development);
2. versioned release milestones, in version order;
3. future/unversioned candidates and deferred work.

Do not assign a version number to an unversioned track until its scope is accepted into a release. When a version is assigned, it must be the next semantic version that matches the scope (major/minor/patch), not a placeholder.

Local Blender skills (development aid)

The MIT blender-skills pack is installed locally at .agents/skills/blender-skills/ (94 skills plus a shared references/ folder). It is gitignored on purpose: it is not game content, not part of the build, tests, CI or release packages, and every contributor may or may not have it. Pi gates project .agents/skills behind project trust, so the trust prompt has to be accepted before the skills load; /skill:blender-director forces one.

Use them when a task touches tools/blender_assets.py, tools/npc_pipeline.py, tools/bake_ao.py, assets/models/blender/**, or Blender-authored venue/prop geometry.

Routing

Start with blender-director for anything multi-discipline, then load only the specialists the task needs:

Work	Skills
Plan/route a Blender task, reference photo match	blender-director
Character mesh, face, hands, proportions	character-artist, character-archetypes, retopology
Skeleton, weights, pose/contract rigging	rigging
UV layout, texel density, UDIM	uv-workflow
Bakery, atlas, normal/AO/ORM maps	texture-workflow
PBR materials and shading	materials
Screenshot-driven material/light iteration	lookdev
Triangle/vertex budgets, naming, scene hygiene	asset-optimization
LOD tiers and impostors	lod-pipeline
Collision shapes and proxies	collision-proxy
GLB export for Godot	godot-export, export-pipeline
Gear, props, street furniture	prop-artist, hard-surface
Trees, hedges, foliage cards	vegetation-artist
Facades, venues, modular kits, dressing	environment-artist, archviz, set-dressing, geometry-nodes
Lighting, framing, final render/composite	lighting, camera-cinematography, rendering, compositing
Style direction (stylized / low-poly / realistic)	stylized-style, lowpoly-style, realistic-style
Ship/no-ship review	qa-review

The same skills cover the local map track (procedural geometry, facade kits, OSM massing, splat reference renders) and the NPC roster work.

Project rules that override a skill

1. The project contract wins. npc_base mesh node, the 13 contract bones at their rest positions, five tint slots, front on Blender −Y / Godot +Z, grounded feet, 8,000 triangles / 8,000 glTF vertices per character, 1,500 per gear, 4,000 per prop. A skill may recommend another topology, naming or budget scheme; do not adopt it unless the change is explicitly planned and the contract, tests and docs are updated together.
2. There is no Blender MCP here. The skills assume MCP tools; this project runs Blender as headless flatpak scripts (tools/blender_assets.py, tools/npc_pipeline.py). Translate MCP steps into script changes or CLI runs. Never add an MCP dependency to the build, CI or runtime.
3. CI stays offline and skill-free. Skills are local guidance only; every gate must still work from a fresh checkout with committed .glb/.png/.tres files and Godot.
4. No new asset source without the licence gate. Skills may suggest reference packs or generated meshes; ADR-0015 (CC0, or CC-BY with attribution in CREDITS.md) and the provenance rules in docs/npc-model-sourcing.md still apply.
5. Evidence, not vibes. After a skill-informed change, run blender_assets.py inspect, audit and fit plus the relevant capture/smoke gates, and say in the handoff which skills informed the change and what the measurements were.

Required Workflow

1. Start by Inspecting the Repository

Before making changes, an agent should:

• Read AGENTS.md.
• Read PLAN.md and README.md if they exist.
• Read the relevant sections of TODO.md and MEMORY.md.
• Inspect the repository structure and existing implementation before proposing changes.
• Check the current version-control status when available.

If TODO.md, DONE.md, or MEMORY.md does not exist, create it using the recommended format in this document before continuing.

Do not assume that a task is unimplemented without checking the code and documentation first.

2. Maintain TODO.md

All planned implementation work must be represented in TODO.md.

• Use headings to group related work into sections.
• Use level-2 headings (##) for milestones or stable themes. A milestone should state its goal and an explicit exit criterion; a theme groups small related work that does not need a milestone.
• Order level-2 sections by current priority: unversioned development tracks first, then versioned release milestones in version order, then future/deferred work. Do not assign a version number to an unversioned track until it is scheduled for a release.
• Use level-3 headings (###) for workstreams inside a milestone (for example, 0.8.1 First-shift tutorial). Keep the hierarchy to three levels: milestone → workstream → checkbox.
• Use checkbox items for each concrete task, normally under a level-3 workstream. Direct checkboxes under a level-2 section are acceptable for small, single-workstream sections.
• Keep tasks small enough to complete and review independently.
• Add new tasks when discoveries expand the scope.
• When starting a task, claim it for the agent workflow by changing [ ] to [-]. Keep it [-] while implementation or verification is in progress; only change it to [x] after the work and relevant checks are finished. If work is abandoned or blocked, leave a clear note and return the item to [ ] (or add a follow-up task).
• Mark a task complete only after the implementation and relevant verification are finished.
• If any verification check fails, do not mark the task complete. Fix the failure, or if it cannot be fixed, add a follow-up task to TODO.md, record the failure in MEMORY.md, and report it in the handoff summary.
• Do not use prose status updates as a substitute for checkboxes.
• Revisit TODO.md regularly and mark checkbox items as complete [x] when the corresponding work is finished.

Recommended format:

# TODO

## Milestone 0.8.0 — First hour and content depth

### 0.8.1 First-shift tutorial

- [x] Write the tutorial beat sheet.
- [-] Implement the tutorial state machine. (claimed; implementation/verification in progress)
- [ ] Add the tutorial smoke test.

### 0.8.2 Campaign shell

- [ ] Define campaign data.
- [ ] Persist campaign progress.

**Exit:** a first-time player completes a shift without external documentation.

## Documentation

- [ ] Document the test procedure in `README.md`.

Order the actual file as described in "Priority and Versioning": unversioned development tracks first (for example character and local-map development), then versioned milestones in version order, then future/deferred work.

3. Move Completed Sections to DONE.md

TODO.md uses nested sections. Apply these completion rules:

1. Level-3 workstream: verify the work, mark every checkbox [x], and add a short completion note. Leave the completed workstream under its parent milestone; do not move an isolated level-3 section to DONE.md.
2. Level-2 milestone/theme: it is complete only when its direct checkboxes and all of its level-3 workstreams are complete. Verify the whole milestone, record its capabilities in docs/development-status.md when it represents shipped functionality, and write the matching CHANGELOG.md entry.
3. Move the entire level-2 section, including all level-3 headings and checklists, to DONE.md. Remove it from TODO.md and add a completion date/note.
4. Planning/overview sections such as a gap analysis, roadmap overview, dependencies or deferred ideas are not completion milestones. Do not move them merely because their checkboxes are empty or complete; update or supersede them as the roadmap changes.
5. Do not move partially completed milestones. Leave them in TODO.md until every nested task is complete, or split the unfinished work into a new smaller level-2 milestone first.

When a completed milestone represents shipped functionality, also record the new capability in docs/development-status.md so the implemented-features list stays accurate, and write a matching entry in CHANGELOG.md (see "Maintain CHANGELOG.md" below).

4. Maintain MEMORY.md

Use MEMORY.md for information that should persist between sessions but does not belong in the general plan or task list.

Record items such as:

• Important design and architecture decisions
• Notable implementation state and gotchas (the canonical implemented-features list lives in docs/development-status.md, not here)
• Known bugs and workarounds
• Commands needed to run, test, or build the project
• Decisions that future agents should not revisit unnecessarily
• Open questions and assumptions
• Files or systems that need special care

Keep memory concise and factual. Remove obsolete entries instead of allowing contradictory notes to accumulate.

Recommended format:

# MEMORY

## Decisions

- The emergency response is a rare last-resort system, not a general combat mechanic.

## Current State

- The game-design document exists in `PLAN.md`.

## Open Questions

- Choose the game engine before implementing the prototype.

5. Maintain PLAN.md

PLAN.md describes what the project should achieve at a high level. It should contain the intended player experience, major systems, scope, and design constraints.

• Update it when the project direction changes.
• Avoid using it as a detailed progress log.
• Convert actionable implementation work from the plan into tasks in TODO.md.
• Keep the plan consistent with decisions recorded in MEMORY.md.

6. Maintain the Root README.md

The root-level README.md is the project’s public-facing technical and project overview. It should document:

• The game’s premise and current status
• The planned or implemented features
• Required tools and dependencies
• Setup instructions
• Run, build, and test commands
• Basic project structure
• Development workflow
• Links to PLAN.md, TODO.md, DONE.md, MEMORY.md, and this file where appropriate
• Known limitations and contribution guidance

Update README.md whenever setup instructions, commands, project structure, or user-facing behavior changes. The README is intentionally concise and links to the docs/ files; keep docs/development-status.md (and the other docs it links to) synchronized with the implementation rather than pasting detail back into the README.

7. Maintain CHANGELOG.md

The root-level CHANGELOG.md records the user-facing changes for every release. It exists so players and contributors can see what changed between versions without reading the commit log.

• Add an entry for every patch, minor, and major version change.
• Follow Semantic Versioning (major.minor.patch): bump major for breaking or vision-level changes, minor for new features, and patch for fixes and small polish.
• A version increment is not complete until its CHANGELOG.md entry, config/version update and matching local release build have all been produced and verified.
• Group each entry under Added, Changed, Fixed, and Removed headings, and date every entry.
• Describe changes in user-facing terms, not as raw commit messages.
• Add the entry when its work ships. If the version number is not yet known, record it under Unreleased and move it into a numbered version when the version is tagged.

When a full executable level-2 milestone/theme in TODO.md is implemented, completed, and moved to DONE.md:

1. Verify the work and mark every checkbox in the section [x].
2. Use the section heading and its completed items to describe the changes and write a CHANGELOG.md entry under the next appropriate version.
3. Move the section to DONE.md.

Do not move a completed section to DONE.md without adding its CHANGELOG.md entry.

8. Implement in Small Increments

For each task:

1. Select or create a task in TODO.md.
2. Inspect the relevant files and dependencies.
3. Make the smallest coherent change that completes the task.
4. Run the relevant tests, checks, or manual verification.
5. Update documentation if the change creates lasting context: add features to docs/development-status.md, adjust the README links or other docs/ files as needed, and record decisions/gotchas in MEMORY.md.
6. Mark the task complete.
7. Keep completed level-3 workstreams in place until the parent milestone is complete; then move the entire level-2 milestone to DONE.md after recording its CHANGELOG.md entry.
8. Commit each coherent, verified change as it is completed. Do not wait until a whole milestone is finished; keep commits small and descriptive and push them when the branch is shareable.
9. On every major, minor or patch version increment: update config/version, add the CHANGELOG.md entry, run the local release build (tools/build_releases.sh), verify the artifacts and checksums, then commit, tag and push the version.
10. If a level-2 milestone from TODO.md has been fully completed, ensure its CHANGELOG.md entry and matching build are in place, then create the next semantic version or tag.
11. Push committed changes to the shared repository if applicable.
12. Note the new version, build artifacts or tag in the relevant documentation and handoff summary.

Avoid unrelated refactors unless they are required for the task or explicitly approved.

9. Commit, Build and Version Discipline

• Commit often. Make a commit after each coherent change or verified task, not only when a milestone or release is finished. Keep the working tree clean between tasks where possible.
• Push regularly. Push shareable commits to the shared repository; do not accumulate a long local-only history.
• Build every version. Major, minor and patch increments each require a local release build and artifact/checksum verification before tagging. A changelog entry alone is not a release.
• CI is a second gate, not a substitute. Tag builds remain automated, but the local build is the required pre-tag check for every version.
• Unversioned tracks stay unversioned. Character, map and other development tracks do not get a placeholder version; the version is assigned when the work is scoped into a release.

Collaboration Rules

• Prefer clear, reversible changes over broad rewrites.
• Preserve existing behavior unless the task requires changing it.
• Explain assumptions when requirements are ambiguous.
• Do not claim a task is complete without verification.
• Keep documentation synchronized with the implementation.
• Mention files changed and checks run in handoff summaries.
• Commit coherent increments as you go; do not wait until the end of a milestone.
• Run a release build whenever a major, minor or patch version is incremented.
• Leave follow-up work as checkbox items in TODO.md rather than hiding it in prose.
• If blocked, document the blocker in MEMORY.md and add a corresponding task to TODO.md.
• Avoid overwriting another contributor’s work; inspect recent changes before editing shared files.

Verification Expectations

Before handing off work, an agent must run every check below that applies to the changed files. Automated tests and formatting/linting/type checks apply to any change to source code. A smoke test applies to any change that affects runtime behavior. Documentation-only changes require only the documentation check. Skipped checks must be listed with a reason.

• Run automated tests.
• Run formatting, linting, or type checks.
• Start the game or application and perform a basic smoke test.
• For 3D/Blender changes: blender_assets.py inspect, audit and fit, the relevant capture gate, and a note of which local skills informed the change.
• Check that documentation and task status reflect the actual state.
• Review the final diff for accidental changes or unfinished markers.

If a check cannot be run, state why and record the missing verification in MEMORY.md or TODO.md.

Handoff Format

A useful handoff should include:

• A brief summary of what changed
• Files added or modified
• Tests or checks that were run
• Any known issues or unverified behavior
• The next recommended task from TODO.md

Safety and Design Constraints

The game should emphasize observation, communication, de-escalation, and proportionate responses. Emergency assistance and weapons must remain rare last-resort mechanics with meaningful consequences, not default solutions for ordinary NPC behavior.