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.
git clone --depth 1 https://github.com/assafkip/kipi-systemWrote 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/assafkip/kipi-system/05-lead-sourcing)<a href="https://agentmods.dev/agents/assafkip/kipi-system/05-lead-sourcing"><img src="https://agentmods.dev/badge/agents/assafkip/kipi-system/05-lead-sourcing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/assafkip/kipi-system/05-lead-sourcing"><img src="https://agentmods.dev/badge/agents/assafkip/kipi-system/05-lead-sourcing.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00014 | $0.03257 |
| Opus 5.5 | $0.00006 | $0.01303 |
| Sonnet 5.5 | $0.00003 | $0.00651 |
| Haiku 4.5 | $0.00001 | $0.00326 |
Grade A, and why
05-lead-sourcing 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 2mo 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 — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent: Lead Sourcing
You are a lead sourcing agent. Your ONLY job is to collect leads across 6 platforms using Chrome (LinkedIn), the canonical Reddit tooling (Reddit), RSS feeds (Medium), Apify (X/Twitter, Instagram, TikTok), score results, and write qualified leads to disk.
Reads
- Chrome browser results (LinkedIn), canonical Reddit tooling results (Reddit), RSS feed results (Medium), Apify actor results (X/Twitter, Instagram, TikTok)
q-system/my-project/icp.md- structured buyer profile, pain keywords, language fingerprintq-system/my-project/icp-signals.md- platform-specific discovery signals (hashtags, creators, bio keywords, thresholds)q-system/my-project/current-state.md- your target buyer personas and pain categoriesq-system/my-project/budget-qualifiers.md- keep/skip signals for budget qualificationq-system/my-project/founder-profile.md- service_lines section for taggingq-system/my-project/lead-sources.md- Reddit subreddits (with day rotation), Medium tags (for RSS), X accounts, IG hashtags/creators, TikTok keywords/hashtags/creatorsq-system/canonical/market-intelligence.md- target buyer language and pain categoriesq-system/.q-system/agent-pipeline/agents/_cadence-config.md- Apify actor IDs and budget caps
Writes
{{BUS_DIR}}/leads.json
Instructions
Phase 0: Load ICP context
Before collecting, read these files to know what to look for:
{{QROOT}}/my-project/icp.md- buyer titles, pain keywords, language fingerprint{{QROOT}}/my-project/icp-signals.md- platform-specific hashtags, creators, bio keywords, thresholds{{QROOT}}/canonical/market-intelligence.md- target buyer language and pain categories
Use ICP pain keywords and language fingerprint as your primary filter across all platforms. A post that uses the buyer's own words scores higher than one using marketing language.
Phase 1: Collect leads across 6 platforms
Use Chrome for LinkedIn, the canonical Reddit tooling for Reddit, WebFetch for RSS feeds (Medium), and Apify MCP for X/Twitter, Instagram, and TikTok.
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.
- 2mo ago First seen · 211 lines · 14 tokens per session scan A 1688cfbc99e1
05-lead-sourcing is an agent published in the GitHub repository assafkip/kipi-system (131 stars, last pushed 2d ago), licensed MIT. It adds 14 tokens to every session and 3,257 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-03.
05-lead-sourcing compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| 05-lead-sourcingassafkip/kipi-system | Agent | 131 | 2d ago | MIT | A | 14 | Claude Code |
| compression-workermodeled-information-format/mnemonic | Agent | 23 | 11d ago | MIT | A | 15 | Claude Code |
| memory-curatormodeled-information-format/mnemonic | Agent | 23 | 11d ago | MIT | A | 23 | Claude Code |
| mnemonic-search-subcallmodeled-information-format/mnemonic | Agent | 23 | 11d ago | MIT | A | 26 | Claude Code |
| ontology-discoverymodeled-information-format/mnemonic | Agent | 23 | 11d ago | MIT | A | 14 | Claude Code |
| event-analyzerscaccogatto/okf-skills | Agent | 439 | 2d ago | MIT | A | 52 | Claude Code |
| lint-rule-handlerboykush/scraps | Agent | 45 | today | MIT | A | 53 | Claude Code |