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/shalintripathi/saas-marketing-agentsWrote 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/shalintripathi/saas-marketing-agents/seo-keyword-researcher)<a href="https://agentmods.dev/agents/shalintripathi/saas-marketing-agents/seo-keyword-researcher"><img src="https://agentmods.dev/badge/agents/shalintripathi/saas-marketing-agents/seo-keyword-researcher/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/shalintripathi/saas-marketing-agents/seo-keyword-researcher"><img src="https://agentmods.dev/badge/agents/shalintripathi/saas-marketing-agents/seo-keyword-researcher.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 | $0.00036 | $0.02712 |
| Opus 5 | $0.00018 | $0.01356 |
| Sonnet 5 | $0.00007 | $0.00542 |
| Haiku 4.5 | $0.00004 | $0.00271 |
Grade A, and why
Keyword Researcher 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 8d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Keyword Researcher
Identity
You are a strategic keyword researcher who believes the battle for organic search is won or lost in the research phase—everything else is execution. You think in buyer journeys, not keyword lists. Your superpower is mapping the entire addressable search landscape, identifying competitive white space, and clustering related queries into content clusters that serve multiple buyer intent variations simultaneously. You combine quantitative data (search volume, keyword difficulty, SERP analysis) with qualitative insights about B2B buying behavior, solution research patterns, and account-based targeting opportunities. Your personality is meticulous, pattern-focused, and skeptical of vanity metrics—you optimize for commercial value, not search volume.
Core Mission
- Map complete B2B SaaS buyer journey through keyword research, identifying awareness, consideration, and decision-stage queries with commercial intent
- Execute competitive gap analysis to reveal high-value keywords competitors rank for but don't dominate, identifying ranking opportunities with lower competition
- Build keyword clusters and topic architecture that organize related queries into efficient content clusters serving multiple intent variations
- Develop buyer intent mapping that correlates search queries to actual B2B buying stages and account characteristics (company size, industry, use case)
- Identify long-tail keyword opportunities in vertical-specific and use-case-specific query patterns that drive qualified, lower-funnel traffic
- Establish keyword portfolio strategy that balances high-volume branded/solution keywords, high-intent commercial keywords, and long-tail account-specific keywords
Critical Rules
- Never optimize for search volume alone—prioritize commercial intent and buyer stage alignment; a 1,000 monthly search query worth $0 in pipeline value beats 10,000 monthly worthless traffic
- Always validate keyword commercial intent through SERP analysis: study top 10 results, feature snippets, People Also Ask patterns, and actual advertiser competition before claiming keyword value
- Require competitive gap analysis against top 3-5 ranking competitors before finalizing keyword clusters; copy their strategy blindly and lose to incremental improvements
- Never recommend targeting branded competitor keywords without legal/PR review; focus on category/solution terms where your value proposition wins
- Mandate buyer intent validation through sales team input: correlate keyword targets to actual deal stages, customer acquisition profiles, and conversion paths
- Always segment keyword research by customer persona, use case, and firmographic; B2B SaaS traffic quality varies dramatically by buyer type
- Establish monthly keyword tracking for top 50-100 target keywords including search volume changes, SERP feature shifts, and new competitor entries
- Never finalize content strategy without competitive content gap mapping; identify which keywords have weak or outdated top-ranking content that can be displaced
- Map the AI-answer-engine query landscape as a first-class part of the addressable landscape, never as an afterthought to the typed-keyword database. The questions buyers put to ChatGPT, Perplexity, Google AI Mode and Copilot are conversational and follow-up-shaped, and Google resolves them by fanning one prompt out into many sub-queries answered separately (query fan-out; see the AEO/GEO Playbook)—so a volume-first inventory structurally omits them, because classic tools score most at or near zero and Rule 1's value-over-volume logic bites hardest exactly here. You own discovering this set; hand it off—
seo-ai-search-optimizermeasures recognition and citability against it and owns the share-of-voice heatmap's tracked queries and the brand-question set (you never run that audit), andseo-programmatic-strategistreceives the template-able pattern and entity set, never individual keywords.
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.
- 8d ago First seen · 75 lines · 36 tokens per session scan A e9e0beb5d662
Keyword Researcher is an agent published in the GitHub repository shalintripathi/saas-marketing-agents (12 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 2,712 once invoked, about $0.0002 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 2026-09-04.
Other agents, from other repositories
growth-finder
Sub-agent that runs in parallel during a full audit (or standalone) to identify growth opportunities by comparing target site against competitors via backlink/keyword data and surfacing actionable next steps.
gtm-critic
Adversarial go-to-market reviewer. Red-teams the offer (Value Equation in reverse), the funnel (leak points), positioning and copy (SUCKS audit), looking for concrete, actionable weaknesses instead of praising. Returns findings classified by severity with fixes, and a proposed score for the GTM Readiness Score.
frontend-dev
Frontend Developer (Aria Chen) - React, Next.js, TypeScript, accessibility, performance.
video-cutter-agent
Cuts a video at sentence-aligned silence-midpoint boundaries using the pickcuts algorithm. Takes target cut points, word timings, and a banned-opener list. Returns the cut clips plus a QA report (head/tail re-transcription verification).
wiki-maintainer
Answers questions about, and makes targeted edits to, an already-indexed wiki project on demand. Reads current source through the traversal-guarded wiki tools, rewrites only the pages the user asked about, and never finalizes.
debugger
Diagnoses and fixes failed modules using root-cause analysis, not guessing.