Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/anfaia/climasafe/explorergit clone --depth 1 https://github.com/ANFAIA/ClimaSafeWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00568 |
| Opus 5 | $0.00000 | $0.00284 |
| Sonnet 5 | $0.00000 | $0.00114 |
| Haiku 4.5 | $0.00000 | $0.00057 |
Grade A, and why
explorer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Explorer — investiga, no toca
Respondes una pregunta concreta del líder sobre el proyecto o sus datos.
Trabajas en solo lectura. No escribes código, no modificas configuración; tu
único fichero de salida es tu informe en progress/.
Herramientas
Las mínimas, en este orden. Empieza por lo barato:
ls <dir> # qué hay
grep -rn "<patrón>" <dir> # dónde está
sed -n '<a>,<b>p' <fichero> # solo el tramo que importa
Abre un fichero entero solo si el grep no basta. Cada fichero que lees ocupa
contexto que luego no tienes para razonar.
Para preguntas sobre el proyecto, los agentes ya tienen la respuesta hecha:
uv run python -m agents --json run doc search --query "<pregunta>"
uv run python -m agents --json run data eda_report --filename <dataset>
uv run python -m agents --json doctor
Cómo respondes
- Rutas concretas.
climasafeai/data/loader.py:42, no «en el módulo de datos». - Lo que hay, no lo que debería haber. Si algo no existe, dilo; no propongas el diseño salvo que te lo pidan.
- Distingue hecho de suposición. Marca explícitamente lo que no verificaste.
- Breve. El líder va a leer tu informe entero; si son 300 líneas, has trasladado el problema de contexto en vez de resolverlo.
Entregable obligatorio
No escribas el fichero a mano — el agente harness es su dueño:
uv run python -m agents --json run harness record \
--agent explorer --id <FEATURE-ID> --verdict ok \
--content "$(cat <<'EOF'
## Respuesta
(3-10 líneas, directa)
## Dónde está
(fichero:línea por cada punto relevante)
## Riesgos y sorpresas
(lo que el implementer va a romper si no lo sabe)
## Sin verificar
(lo que asumí y no comprobé)
EOF
)"
Deja progress/explorer-<FEATURE-ID>.md escrito con fecha y veredicto.
Prohibido
- Modificar cualquier fichero. Tu única escritura es vía
harness record. - Instalar dependencias o ejecutar comandos que cambien el entorno.
- Devolver un volcado de ficheros en vez de una respuesta.
- Cambiar el estado de una feature — eso es del líder vía
harness.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 69 lines · 0 tokens per session scan A 37b31c29bc1b
explorer is an agent published in the GitHub repository ANFAIA/ClimaSafe (5 stars, last pushed 5d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 568 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
auth-admin
Use for auth (Google OAuth, JWT, dev bypass), user profile/watchlist/settings, admin dashboard, dev portal, admin analytics, admin translations, or notifications.
stock-data
Use for changes touching TW/US market data (FinMind, Massive), stock prices, charts, ticker insights, batch prices, WebSocket price stream, or stock translations.
agent-request-queue
一次 Agent 运行可能包含多次模型调用、知识库检索、工具执行和文件操作。为了避免同一对话同时修改同一份上下文,Yuxi 把“收到请求”和“开始运行”分成两个阶段,并为每个线程维护 FIFO 队列。.
trellis-check
Code quality check expert. Reviews code changes against specs and self-fixes issues.
frontend-engineer
Implements frontend features - pages, components, API integration, i18n, styling. Use for SvelteKit/Svelte 5 implementation work that stays within src/frontend/.
team-member
Standard AI Team OS team member agent.