upstream/K-Dense-AI-scientific-agent-skills/catalogue/AGENTS.md

@@ -2,9 +2,9 @@
 title: "Repository Guidance"
 task: ""
 lineage_type: import
-upstream_source: https://github.com/K-Dense-AI/scientific-agent-skills/blob/991bd993/AGENTS.md
+upstream_source: https://github.com/K-Dense-AI/scientific-agent-skills/blob/de66e10c/AGENTS.md
-upstream_sha: 991bd993
+upstream_sha: de66e10c
-imported_at: 2026-08-08
+imported_at: 2026-08-19
 prompt_class: unknown
 upstream_changes: accepted
 author: upstream
@@ -38,7 +38,14 @@ The general-purpose skills that do exist are narrow output-format helpers (`docx
 
 ## Layout
 
+The repository root is an [Agent Plugins](https://agent-plugins.org/) 1.0.0 package: `plugin.json`
+plus the portable `skills/` tree. Keep `plugin.json` valid against the Agent Plugins manifest
+schema, and keep its `version` identical to `pyproject.toml` `[project].version`. Do not add
+non-portable top-level fields to `plugin.json` (no inline MCP, hooks, or client-only keys — use
+`mcp.json` or a reverse-domain `extensions` namespace if those are ever needed).
+
 ```text
+plugin.json                 # Agent Plugins manifest (repo root)
 skills/<skill-name>/
 ├── SKILL.md        # required
 ├── references/     # optional: long documentation, loaded only when needed
@@ -46,8 +53,8 @@ skills/<skill-name>/
 └── assets/         # optional: templates and static resources
 ```
 
-Only `SKILL.md` is required. Reference other files with relative paths from the skill root, kept
+Only `SKILL.md` is required inside each skill. Reference other files with relative paths from the
-one level deep.
+skill root, kept one level deep.
 
 **Tests never live under `skills/`.** A skill directory ships only what an agent loads. Checks for a
 skill's scripts and structure go in the repository-level suite instead:
@@ -58,9 +65,9 @@ tests/<skill-name>/          # same name as the skill directory
 └── fixtures/                # optional test data
 ```
 
-**Diagrams never live under `skills/` either.** Every skill has one generated workflow diagram at
+**Diagrams never live under `skills/` either.** A skill may have a generated workflow diagram at
-`docs/images/<skill-name>.png`, produced by `scripts/generate_skill_image.py` and kept in step with
+`docs/images/<skill-name>.png`, produced by `scripts/generate_skill_image.py`. Diagrams are optional
-the skill's documentation — see [Skill diagrams](#skill-diagrams).
+— see [Skill diagrams](#skill-diagrams).
 
 Tests reach their skill through an explicit anchor, never a relative walk:
 
@@ -79,11 +86,6 @@ SKILL_ROOT = Path(__file__).resolve().parents[2] / "skills" / "<skill-name>"
 5. If the skill ships `scripts/`, put their tests in **`tests/<name>/`** — never in the skill
    directory. Fixtures go in `tests/<name>/fixtures/`.
 6. Validate and scan (below).
-7. Generate the skill's diagram — a new skill without `docs/images/<name>.png` is incomplete:
-
-   ```bash
-   uv run python scripts/generate_skill_image.py --skill <name>
-   ```
 
 ```markdown
 ---
@@ -121,16 +123,6 @@ Use this skill when...
 5. Re-run any example, command, or script you touched, plus `tests/<name>/` if that suite exists.
    Suites check that `metadata.version` is present and quoted, not what it equals, so a version bump
    never needs a matching test edit.
-6. **Regenerate the diagram in the same change** whenever the edit changes what the skill does or
-   how its workflow runs — the picture is generated from `SKILL.md` and `references/`, so it goes
-   stale silently. The command overwrites `docs/images/<name>.png` in place:
-
-   ```bash
-   uv run python scripts/generate_skill_image.py --skill <name>
-   ```
-
-   A typo fix, a link repair, or a version bump alone does not need a new image.
-
 ## Frontmatter
 
 `SKILL.md` starts with YAML frontmatter. **Only these six fields are allowed** — the spec defines a
@@ -337,10 +329,11 @@ touch the shared contract.
 
 ## Skill diagrams
 
-Every skill carries one generated workflow diagram at `docs/images/<skill-name>.png`. Creating a
+A skill may carry a generated workflow diagram at `docs/images/<skill-name>.png`. Diagrams are
-skill means creating its image; changing what a skill does means regenerating it. The image is not
+optional: neither a new skill nor a change to an existing one is blocked on having or refreshing an
-optional decoration — it is derived from the documentation, so an out-of-date one misrepresents the
+image, and no CI check enforces them. If you do ship one, note that it is derived from the
-skill.
+documentation, so regenerate it when the skill's workflow changes rather than leaving a picture that
+misrepresents the skill.
 
 `scripts/generate_skill_image.py` is local repository tooling, standard library only, and runs in
 two stages on one `OPENROUTER_API_KEY` (environment variable, repository `.env`, or `--api-key`):
@@ -380,12 +373,13 @@ hand-tuning one skill's prompt, so the set stays visually consistent.
 - `metadata.version` exists, is quoted, and is bumped if you changed an existing skill.
 - `metadata` is a block mapping; `openclaw` / `hermes` blocks are nested mappings.
 - `uv run skills-ref validate skills/<name>` passes.
+- If the collection version changes, `plugin.json` `version` matches `pyproject.toml`.
 - `uv run --with pytest python -m pytest tests/_meta -q` passes — this is what CI blocks on, and it
-  catches a missing suite, a missing `skill-requirements.toml` entry, a broken local link, and a
+  catches a missing suite, a missing `skill-requirements.toml` entry, a broken local link, a
-  leaked local path.
+  leaked local path, and a drifted Agent Plugins manifest.
 - If the skill ships `scripts/`: a suite exists at `tests/<name>/`, a `[skills.<name>]` entry exists
   in `tests/skill-requirements.toml`, and `python tests/run_all.py --isolated <name>` passes.
-- `docs/images/<name>.png` exists, and was regenerated if the change altered what the skill does.
+- If the skill ships `docs/images/<name>.png`, its labels are spelled correctly and its arrows point
-  Its labels are spelled correctly and its arrows point where they should.
+  where they should. The image itself is optional.
 - Examples and scripts are tested, or clearly marked illustrative.
 - No secrets or private data; scan results clean or explained in the PR.