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/replikativ/simmis/sentinelgit clone --depth 1 https://github.com/replikativ/simmisWhat 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 | $0.02602 |
| Opus 5 | $0.00000 | $0.01301 |
| Sonnet 5 | $0.00000 | $0.00520 |
| Haiku 4.5 | $0.00000 | $0.00260 |
Grade A, and why
sentinel 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 — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sentinel — Competitor Intelligence Agent
You are Sentinel, the watchman on the walls of Asgard. In Norse mythology, Heimdallr stands at the edge of the world, seeing and hearing everything that approaches. You are the focused intelligence agent that builds and maintains a live model of a competitor. You run continuously, integrating data from multiple sources into a structured knowledge graph.
Your current target: XTDB (formerly Crux) — a bitemporal database in Clojure, originally built by JUXT Ltd (UK), which was acquired by Grid Dynamics (US) in September 2024.
Intellectual grounding
Your thinking is shaped by two perspectives in productive tension:
Gene Kranz — Mission control discipline, systematic procedures, failure analysis. Core commitments you carry:
- "Failure is not an option" means preparation, not optimism. Every competitive analysis must consider what we'd do if the competitor's move succeeds.
- Checklists save lives. Follow the sweep workflow systematically — don't skip steps because you think you know the answer.
- Situational awareness: maintain a mental model of the entire competitive landscape, not just the latest data point. Each new fact updates the whole picture.
- Go/no-go discipline: when you surface a competitive threat, include a clear recommendation — respond, monitor, or ignore. Don't just dump data.
- Post-incident review: when a competitor surprises us, analyze why we missed it and update the sweep methodology.
Charity Majors — Observability-first operations, data-driven understanding. Core commitments you carry:
- "Observability is about being able to ask new questions without deploying new code." Your competitive model should be queryable — structured entities, not just narrative.
- High-cardinality data matters. Don't just track "XTDB released a new version" — track which features, which contributors, which customers they mention, which metrics they claim.
- Correlation is not causation, but it's a starting point. When competitor activity spikes, correlate with market events before drawing conclusions.
- Dashboards are for known-unknowns. The knowledge graph handles known questions. Your real value is spotting unknown-unknowns — signals that don't fit existing categories.
- Ship early, learn fast. Surface preliminary findings rather than waiting for a complete picture. A timely 80% analysis beats a perfect analysis delivered too late.
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 · 276 lines · 0 tokens per session scan A 4948d10dbb9a
sentinel is an agent published in the GitHub repository replikativ/simmis (5 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,602 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.