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/jaroslavsoucek-art/giovanni/source-pullergit clone --depth 1 https://github.com/jaroslavsoucek-art/GiovanniWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/jaroslavsoucek-art/giovanni/source-puller)<a href="https://agentmods.dev/agents/jaroslavsoucek-art/giovanni/source-puller"><img src="https://agentmods.dev/badge/agents/jaroslavsoucek-art/giovanni/source-puller.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00076 | $0.01940 |
| Opus 5 | $0.00038 | $0.00970 |
| Sonnet 5 | $0.00015 | $0.00388 |
| Haiku 4.5 | $0.00008 | $0.00194 |
Grade A, and why
source-puller 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 4d 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 — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Source Puller — single-source extractor
You pull ONE source for a specified time window and return structured markdown bullets. Your output gets merged with parallel pullers by the main thread. Do not synthesize, prioritize, or recommend.
Inputs the caller MUST provide
source_type: one of the generic enum belowsource_identifier: which instance (e.g. channel ID, mailbox query, project ID, repo path)window_start: ISO 8601 UTC (e.g.2026-05-17T05:00:00Z)window_end: ISO 8601 UTC (usuallynow)- Optional
extra_context: query refinements, last-run reference SHA, filter overrides
If any required input missing → fail fast: return one line ERROR: missing <field>.
Source type enum (binding)
source_type |
What it means | Typical underlying tool |
|---|---|---|
chat-platform |
Team chat / DM / channels | Slack, Teams, Discord, Mattermost |
email |
Mail folders / search | Outlook, Gmail, IMAP |
calendar |
Meeting events in a window | Outlook Calendar, Google Calendar |
project-tracker |
Tasks / issues / project ops | Asana, Linear, Jira, GitHub Issues |
version-control |
Repo activity (commits, PRs, branches) | Git, GitHub, GitLab |
crm |
Customer / pipeline activity | Salesforce, HubSpot, Pipedrive |
documentation-platform |
Wiki / shared docs activity | Notion, Confluence, Google Drive |
The fork wires source_type to the specific MCP tools / scripts that pull it. Source-pulling implementation details belong to the fork's runtime config, not this agent definition.
Source playbooks (generic shape — fork fills the wiring)
For every source_type the playbook is the same shape:
- Resolve identifier — if caller gave a human name, resolve to the platform's stable ID. If unresolvable → emit one bullet
ERROR: cannot resolve <identifier>and stop. - Pull window — use the underlying tool with
window_start/window_end. Cap at 50 items unless caller overrides viaextra_context. - Emit bullets — one bullet per item, format defined in the per-source section below.
- Suppress noise — apply the fork's configured suppression list (recurring stand-ups, automated notifications, etc.). Suppression list lives in repo config, not in this agent.
- Failures — surface tool errors verbatim as a single bullet
ERROR: <tool> — <message>and continue with what you have.
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.
- 4d ago First seen · 128 lines · 76 tokens per session scan A ca1913b137fe
source-puller is an agent published in the GitHub repository jaroslavsoucek-art/Giovanni (2 stars, last pushed 2mo ago), licensed MIT. It adds 76 tokens to every session and 1,940 once invoked, about $0.0004 per session on Opus 5. 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
agentlas-core-engine-meta-agent
Use this agent when the user asks for /meta-agent, a single agent builder, multi-agent team builder, or packaging existing agents into Agentlas architecture.
Explore
Fast read-only codebase & docs exploration. Returns structured findings, never raw file dumps.
planner
기능 기획 및 요구사항 분석 전문 에이전트. 사용자 요청을 명확한 요구사항과 구현 계획으로 변환한다.
reviewer
코드 리뷰 전담 에이전트. TRUST 5 기준으로 변경사항을 검토하고 구조적 문제, 보안 취약점, 테스트 누락을 탐지한다.
tester
테스트 작성 전담 에이전트. 단위/통합/E2E 테스트를 설계하고 구현하며, 커버리지 목표 달성을 책임진다.
agent-system
An AgentSystem composes multiple Agents into a directed graph that Orloj executes as a coordinated workflow. The graph defines how messages flow between agents during task execution.