mirror of
https://github.com/wshobson/agents
synced 2026-06-21 14:13:58 +00:00
98862b56d8
* feat: AGENTS.md canonical context + OpenAI harness-engineering layout Promote AGENTS.md to the committed cross-harness context file (per the agents.md convention and OpenAI's harness-engineering blog). Harness- specific files become thin redirects: - AGENTS.md — canonical, committed (~74 lines, table-of-contents) - CLAUDE.md — `@AGENTS.md` import + Claude-specific addenda - GEMINI.md — Gemini-specific setup only - .gemini/settings.json — redirects Gemini CLI's context to read AGENTS.md - ARCHITECTURE.md — new at root, top-level architectural map - gemini-extension.json — bumps version to 1.7.0, sets contextFileName: AGENTS.md - .gitignore — drops the AGENTS.md entry (file is now committed) Harness support verified: - Codex CLI reads AGENTS.md natively (root → cwd walk, 32 KiB cap) - Cursor 2.5+ reads AGENTS.md natively - OpenCode reads AGENTS.md natively (wins over CLAUDE.md if both exist) - Claude Code: `CLAUDE.md` first line is `@AGENTS.md` (Anthropic's documented interop pattern) - Gemini CLI: `.gemini/settings.json` context.fileName redirect (Gemini doesn't support @-imports) Codex adapter no longer generates AGENTS.md — `emit_global` instead validates the committed file fits Codex's 32 KiB cap and the 150-line table-of-contents convention. Tests updated. Clean-output target no longer touches AGENTS.md. ## Auxiliary files updated for multi-harness reality - `.github/ISSUE_TEMPLATE/bug_report.yml` — dropdown for harness + component path; renames "subagent" → "plugin/agent/skill/command" - `.github/ISSUE_TEMPLATE/feature_request.yml` — scope dropdown covers framework / harness / tooling / docs / CI in addition to components - `.github/ISSUE_TEMPLATE/new_subagent.yml` — relabeled "New Component Proposal" with component-type dropdown (plugin/agent/skill/command/ harness adapter) and cross-harness portability field - `.github/ISSUE_TEMPLATE/config.yml` — links to AGENTS.md, authoring guide, per-harness docs; updated Contributing link to root - `.github/CONTRIBUTING.md` — thin pointer to canonical root CONTRIBUTING.md - `.github/PULL_REQUEST_TEMPLATE.md` — new; scope + affected-harness checklists, test-plan checklist, portability-notes section - CONTRIBUTING.md (root) — updated to reference AGENTS.md / ARCHITECTURE.md - gemini-extension.json — version 1.6.0 → 1.7.0, count fixes, redirects to AGENTS.md as contextFileName ## Code-quality CI New `.github/workflows/code-quality.yml` with three jobs: - `python-lint` — `ruff check`, `ruff format --check`, `ty check` on the adapter framework + plugin-eval. yt-design-extractor.py legacy code excluded. - `markdown-lint` — markdownlint-cli2 against README, AGENTS, ARCHITECTURE, CLAUDE, top-level guides, and docs/. Config in `.markdownlint.json`. - `json-lint` — validates every JSON / TOML / YAML in the repo (excluding generated trees). Required ty environment config added to plugin-eval/pyproject.toml so `tools.adapters.*` resolves from outside the package. Fixed one ty error in `tools/adapters/base.py:HarnessAdapter.capabilities` (return-type annotation didn't match the `Capability` dataclass returned). Fixed two ruff SIM108 ternary suggestions in codex.py and doc_gardener.py. ruff format applied across all in-scope files (formatting-only diffs). ## Tests + verification - 387 pytest tests pass (1 new test for the AGENTS.md validate-don't-overwrite behavior) - `make validate STRICT=1` clean - `make garden` 0 errors (10 warnings — remaining oversize source skills) - `make smoke-test` clean against locally installed OpenCode/Gemini/Codex/Claude Code - Real-CLI round-trip: `opencode agent list` discovers 193 subagents, `gemini extensions validate .` succeeds, all 191 Codex agent TOMLs parse ## Tag recommendations (separate task — for repo About panel) Top 20 by reach + relevance (from `gh api search/repositories?q=topic:<tag>`): automation mcp ai-agents developer-tools claude-code anthropic agentic-ai agents prompt-engineering cursor multi-agent agent-skills orchestration opencode workflows gemini-cli codex-cli claude-code-skills cursor-rules claude-code-plugins * fix(ci): YAML multi-doc + markdownlint scope/rules Two CI failures on PR #542, both fixed: ## JSON/TOML/YAML syntax job (3 false-positive YAML errors) The job's YAML validation used `yaml.safe_load` which only reads the first document in a multi-document YAML stream. Three Kubernetes manifest templates use the standard `---` document separator (valid YAML) and were mis-flagged: plugins/kubernetes-operations/skills/k8s-manifest-generator/assets/configmap-template.yaml plugins/kubernetes-operations/skills/k8s-manifest-generator/assets/service-template.yaml plugins/kubernetes-operations/skills/k8s-security-policies/assets/network-policy-template.yaml Switched to `list(yaml.safe_load_all(...))` so multi-doc YAML is accepted. ## Markdown lint job (lots of pre-existing plugin-README violations) Two changes: 1. **Narrow the lint glob** — markdownlint now runs against top-level guides (README, AGENTS, ARCHITECTURE, CLAUDE, per-harness setup, CONTRIBUTING) and our authored `docs/` only. Per-plugin READMEs (`plugins/*/README.md`) are owned by their plugin authors and not lint-gated as part of this framework PR. Lint enforcement for those belongs at the plugin-author layer, not the framework PR layer. 2. **Tighten `.markdownlint.json`** — disable two rules that produce noise without catching real defects: - MD040 (fenced-code-language) — terminal output / shell command blocks frequently omit a language by convention - MD060 (table-column-style) — cosmetic table-pipe spacing; doesn't affect rendering Genuine formatting rules kept: MD029 (ol-prefix), MD031 (blanks- around-fences), MD032 (blanks-around-lists), MD056 (table-column- count), MD058 (blanks-around-tables). ## Real defects caught and fixed The narrower scope still caught 5 real issues: - `docs/agent-skills.md:397` — code fence inside an ordered list item needed a blank line before the fence - `docs/authoring.md:90` — bulleted list needed a blank line above - `docs/plugin-eval.md:87` — table needed a blank line above - `OPENCODE.md:39` — table-column-count error caused by literal `|` inside backticks: ``mode: primary|subagent|all`` (3 cells reads as 5) Rewrote as ``mode:` one of `primary` / `subagent` / `all`` ## Verification - `npx markdownlint-cli2 "*.md" "docs/*.md"` → 0 errors - `yaml.safe_load_all` accepts all multi-doc YAMLs (0 errors) - All other CI jobs already passing (Python ruff/ty, multi-harness generate, CLI smoke test, plugin-eval pytest, tools pytest) * fix: address PR #542 bot feedback - Codex P2 (chatgpt-codex-connector): emit_global now reads AGENTS.md from the repo root (WORKTREE), not output_root. Previously `--output-root <scratch>` produced a false "missing" warning even when AGENTS.md was committed at the real root, breaking --strict generation outside the repo. Added a constructor arg `repo_root` so tests can stage a fake AGENTS.md without touching the committed file, plus a regression test that proves the two paths are decoupled. - CodeRabbit nitpick (code-quality.yml): added workflow-level `permissions: contents: read` and `persist-credentials: false` on every checkout. Skipped the SHA-pinning recommendation — it's a heavier blanket-policy decision and the workflow has no write scope to abuse. - CodeRabbit nitpick (pyproject.toml): consolidated the duplicate `dev` groups by moving `ty` into `[project.optional-dependencies].dev` and removing the now-empty `[dependency-groups]` block. Dropped `--group dev` from `code-quality.yml`'s `uv sync` since `--all-extras` now covers it. - Verified gemini-extension.json counts (82/191/155/102) against the actual source-of-truth: 81 local plugins + 1 external = 82 in marketplace.json, 191 agent .md files, 155 SKILL.md files, 102 command .md files. Counts are correct as-is — CodeRabbit's quick-win was a regex miscount.
217 lines
8.6 KiB
Python
217 lines
8.6 KiB
Python
"""Gemini CLI adapter (refreshed for late 2025 / early 2026 capabilities).
|
|
|
|
Replaces the original `tools/generate_gemini_commands.py`. Three concrete changes
|
|
versus the original:
|
|
|
|
1. Emits native skills at `skills/<plugin>__<skill>/SKILL.md` (Gemini CLI auto-discovers
|
|
these from Dec 2025).
|
|
2. Emits native subagents at `agents/<plugin>__<agent>.md` (April 2026).
|
|
3. Command prompts use Gemini's native `@{path}` file-injection syntax when the body is
|
|
large, OR inline the body directly when it's small enough (~4 KB threshold).
|
|
|
|
Sources: research summary by `a8fba67f1a1ce1db8` synthesized into the plan.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from pathlib import Path
|
|
|
|
from tools.adapters.base import (
|
|
AgentSource,
|
|
CommandSource,
|
|
EmitResult,
|
|
HarnessAdapter,
|
|
PluginSource,
|
|
SkillSource,
|
|
)
|
|
from tools.adapters.capabilities import TOOL_NAME_MAPS, resolve_model
|
|
|
|
_INLINE_BODY_THRESHOLD = 4 * 1024 # bytes; below this, inline; above, use @{path}
|
|
|
|
|
|
def _escape_toml_basic(s: str) -> str:
|
|
return s.replace("\\", "\\\\").replace('"', '\\"')
|
|
|
|
|
|
def _escape_toml_multiline(s: str) -> str:
|
|
return s.replace("\\", "\\\\").replace('"""', '\\"\\"\\"')
|
|
|
|
|
|
def _generate_command_toml(description: str, prompt: str) -> str:
|
|
return (
|
|
f'description = "{_escape_toml_basic(description)}"\n'
|
|
f'prompt = """\n{_escape_toml_multiline(prompt)}\n"""\n'
|
|
)
|
|
|
|
|
|
def _gemini_frontmatter(fm: dict) -> str:
|
|
lines = ["---"]
|
|
for k, v in fm.items():
|
|
if isinstance(v, list):
|
|
value = ", ".join(str(x) for x in v)
|
|
lines.append(f"{k}: [{value}]")
|
|
elif isinstance(v, bool):
|
|
lines.append(f"{k}: {'true' if v else 'false'}")
|
|
elif v is None:
|
|
continue
|
|
else:
|
|
value = str(v).replace("\n", " ").strip()
|
|
lines.append(f"{k}: {value}")
|
|
lines.append("---")
|
|
return "\n".join(lines)
|
|
|
|
|
|
class GeminiAdapter(HarnessAdapter):
|
|
harness_id = "gemini"
|
|
|
|
def emit_plugin(self, plugin: PluginSource) -> EmitResult:
|
|
result = EmitResult()
|
|
for skill in plugin.skills:
|
|
self._emit_skill(plugin, skill, result)
|
|
for agent in plugin.agents:
|
|
self._emit_agent(plugin, agent, result)
|
|
for cmd in plugin.commands:
|
|
self._emit_command(plugin, cmd, result)
|
|
# Plugin entry-point command at commands/<plugin>.toml — matches Gemini's
|
|
# `/<plugin>` top-level namespace convention.
|
|
self._emit_plugin_entry(plugin, result)
|
|
return result
|
|
|
|
# ── Internals ──────────────────────────────────────────────────────────
|
|
|
|
def _emit_skill(self, plugin: PluginSource, skill: SkillSource, result: EmitResult) -> None:
|
|
"""Mirror skill to skills/<plugin>__<skill>/SKILL.md at extension root.
|
|
|
|
Gemini CLI auto-discovers these paths (Dec 2025 SKILL.md spec).
|
|
"""
|
|
skill_id = f"{plugin.name}__{skill.name}"
|
|
rel_dir = Path("skills") / skill_id
|
|
fm = dict(skill.frontmatter)
|
|
fm["name"] = skill_id
|
|
|
|
content = _gemini_frontmatter(fm) + "\n\n" + skill.body.rstrip() + "\n"
|
|
result.written.append(self.write(rel_dir / "SKILL.md", content))
|
|
|
|
# Mirror references/ — binary copy so non-text assets don't crash the run.
|
|
if skill.references_dir:
|
|
for ref in sorted(skill.references_dir.rglob("*")):
|
|
if ref.is_file():
|
|
rel = ref.relative_to(skill.references_dir)
|
|
result.written.append(self.mirror_file(ref, rel_dir / "references" / rel))
|
|
|
|
def _emit_agent(self, plugin: PluginSource, agent: AgentSource, result: EmitResult) -> None:
|
|
"""Emit a Gemini subagent (April 2026 spec) at agents/<plugin>__<agent>.md."""
|
|
agent_id = f"{plugin.name}__{agent.name}"
|
|
rel = Path("agents") / f"{agent_id}.md"
|
|
|
|
model, warning = resolve_model("gemini", agent.model)
|
|
if warning:
|
|
result.warnings.append(f"agent `{agent_id}`: {warning}")
|
|
fm: dict = {
|
|
"name": agent_id,
|
|
"description": agent.description or f"{agent.name} (from {plugin.name})",
|
|
"model": model,
|
|
}
|
|
if agent.tools:
|
|
# Remap Claude Code CamelCase tool names to Gemini's native names.
|
|
# Unmapped tools (e.g. mcp__*, custom names) pass through unchanged.
|
|
gemini_map = TOOL_NAME_MAPS["gemini"]
|
|
fm["tools"] = [gemini_map.get(t, t) for t in agent.tools]
|
|
|
|
content = _gemini_frontmatter(fm) + "\n\n" + agent.body.rstrip() + "\n"
|
|
result.written.append(self.write(rel, content))
|
|
|
|
def _emit_command(self, plugin: PluginSource, cmd: CommandSource, result: EmitResult) -> None:
|
|
"""Emit a Gemini TOML command at commands/<plugin>/<command>.toml.
|
|
|
|
Use `@{path}` file injection for large bodies, inline for small ones (the Google
|
|
security extension's `analyze.toml` is the canonical idiom for inlining).
|
|
"""
|
|
rel = Path("commands") / plugin.name / f"{cmd.name}.toml"
|
|
|
|
description = cmd.description or cmd.name.replace("-", " ").title()
|
|
body_bytes = len(cmd.body.encode("utf-8"))
|
|
|
|
if body_bytes <= _INLINE_BODY_THRESHOLD:
|
|
prompt = self._inline_command_prompt(plugin, cmd, description)
|
|
else:
|
|
prompt = self._inject_command_prompt(plugin, cmd, description)
|
|
|
|
result.written.append(self.write(rel, _generate_command_toml(description, prompt)))
|
|
|
|
def _emit_plugin_entry(self, plugin: PluginSource, result: EmitResult) -> None:
|
|
"""Top-level commands/<plugin>.toml — Gemini exposes this as `/<plugin>`."""
|
|
description = plugin.description or f"{plugin.name.replace('-', ' ').title()} plugin"
|
|
rel = Path("commands") / f"{plugin.name}.toml"
|
|
|
|
agent_list = ", ".join(f"`{plugin.name}__{a.name}`" for a in plugin.agents)
|
|
skill_list = ", ".join(f"`{plugin.name}__{s.name}`" for s in plugin.skills)
|
|
command_list = ", ".join(f"`/{plugin.name}:{c.name}`" for c in plugin.commands)
|
|
|
|
parts = [description.rstrip(".") + "."]
|
|
parts.append("")
|
|
parts.append(f"This is the entry point for the `{plugin.name}` plugin.")
|
|
if plugin.agents:
|
|
parts.append("")
|
|
parts.append(f"Subagents: {agent_list}. Invoke with `@<agent>` syntax.")
|
|
if plugin.skills:
|
|
parts.append("")
|
|
parts.append(f"Skills: {skill_list}. Describe a matching task to activate.")
|
|
if plugin.commands:
|
|
parts.append("")
|
|
parts.append(f"Commands: {command_list}.")
|
|
parts.append("")
|
|
parts.append("{{args}}")
|
|
|
|
result.written.append(
|
|
self.write(rel, _generate_command_toml(description, "\n".join(parts)))
|
|
)
|
|
|
|
def _inline_command_prompt(
|
|
self, plugin: PluginSource, cmd: CommandSource, description: str
|
|
) -> str:
|
|
"""Self-contained prompt with the command body inlined (small commands)."""
|
|
lines = [
|
|
f"You are running the `{cmd.name}` command from the `{plugin.name}` plugin.",
|
|
"",
|
|
"## Protocol",
|
|
"",
|
|
cmd.body.strip(),
|
|
"",
|
|
]
|
|
if cmd.argument_hint:
|
|
lines.append(f"Arguments: {cmd.argument_hint}")
|
|
lines.append("")
|
|
lines.append("{{args}}")
|
|
return "\n".join(lines)
|
|
|
|
def _inject_command_prompt(
|
|
self, plugin: PluginSource, cmd: CommandSource, description: str
|
|
) -> str:
|
|
"""Use @{path} file injection — fixes the broken 'READ the protocol at X' pattern.
|
|
|
|
Gemini resolves @{} paths relative to the extension root at evaluation time.
|
|
"""
|
|
cmd_path = f"plugins/{plugin.name}/commands/{cmd.name}.md"
|
|
lines = [
|
|
f"You are running the `{cmd.name}` command from the `{plugin.name}` plugin.",
|
|
"",
|
|
"## Protocol",
|
|
"",
|
|
"The full protocol definition is injected below. Read it in full before executing.",
|
|
"",
|
|
f"@{{{cmd_path}}}",
|
|
"",
|
|
"## Execution",
|
|
"",
|
|
"1. Initialize the session according to the 'Pre-flight Checks' section in the protocol.",
|
|
"2. Execute the steps sequentially. Pause at every `PHASE CHECKPOINT` / `GATE`.",
|
|
"3. Use available subagents and skills as the protocol directs.",
|
|
"",
|
|
]
|
|
if cmd.argument_hint:
|
|
lines.append(f"Arguments: {cmd.argument_hint}")
|
|
lines.append("")
|
|
lines.append("{{args}}")
|
|
return "\n".join(lines)
|