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 skills add nimrodfisher/agent-think-map --skill nanoclawgit clone --depth 1 https://github.com/nimrodfisher/agent-think-mapWrote 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/skills/nimrodfisher/agent-think-map/nanoclaw)<a href="https://agentmods.dev/skills/nimrodfisher/agent-think-map/nanoclaw"><img src="https://agentmods.dev/badge/skills/nimrodfisher/agent-think-map/nanoclaw/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/skills/nimrodfisher/agent-think-map/nanoclaw"><img src="https://agentmods.dev/badge/skills/nimrodfisher/agent-think-map/nanoclaw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00035 | $0.00689 |
| Opus 5 | $0.00017 | $0.00345 |
| Sonnet 5 | $0.00007 | $0.00138 |
| Haiku 4.5 | $0.00003 | $0.00069 |
Grade A, and why
add-simulator 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 10d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Add Agent Think Map
Install a live thinking canvas into this NanoClaw checkout. NanoClaw has no HTTP API; traces travel as kind: "trace" rows in outbound.db, and the host exposes them as SSE at /webhook/simulator.
Do not merge the channels or providers branches. Copy the files below, append the one-line barrel import, then stop.
Files to copy
Copy from the Agent Think Map repo (this skill's files/ directory) into the NanoClaw checkout:
| Source | Destination |
|---|---|
files/container/agent-runner/src/trace-bridge.ts |
container/agent-runner/src/trace-bridge.ts |
files/src/modules/simulator.ts |
src/modules/simulator.ts |
files/public/simulator.html |
public/simulator.html |
Overwrite if the files already exist (idempotent re-run).
Wire the host module
Append this import to src/modules/index.ts if it is not already present:
import './simulator.js';
If src/modules/index.ts does not exist, append the same line to the host module barrel that already imports scheduling/permissions (search for onHostStart or registerWebhookHandler usages and put the import next to the other module imports).
Wire the runner
In container/agent-runner/src/poll-loop.ts (or the file that calls provider.query()), import and wrap the query iterator:
import { emitTraceFromQuery } from './trace-bridge.ts';
Around the existing provider.query(...) loop, replace the raw iterator with:
const traced = emitTraceFromQuery(provider.query(prompt, options), {
runId: sessionId,
prompt: userText,
writeTrace: (row) => writeMessageOut(row),
});
for await (const message of traced) {
// existing message handling unchanged
}
Enable partial streaming on the query options if the provider is Claude:
options.includePartialMessages = true;
writeMessageOut must persist row.kind === 'trace' into outbound.db messages_out the same way chat rows are written. If the writer rejects unknown kinds, add 'trace' to the allow-list.
What ships with it
8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 80 lines · 35 tokens per session scan A 08d1cc073806
add-simulator is a skill published in the GitHub repository nimrodfisher/agent-think-map (12 stars, last pushed 3d ago), licensed MIT. It adds 35 tokens to every session and 689 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 2026-08-31.
Other skills, from other repositories
agenttrace-session-audit
Audit local AI coding-agent sessions with agenttrace for cost, tool failures, latency, anomalies, health, diffs, and CI gates.
langsmith-observability
LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.
morning-report
Weather + news morning briefing: fetch live weather and headlines, compose a plain-text briefing, deliver it. Multi-step; load before acting.
github-security-fix
Remediate GitHub security / Dependabot / CodeQL / code-quality alerts via fixgithubsecurityalert.
slack-handoff
Connect OpenSRE to Slack and show how to hand off DevOps chores from a channel mention or a DM. Verifies with cliexec; if Slack is missing, queues /integrations setup slack via slashinvoke (the wizard needs a full terminal). Use for the startup demo option "Connect OpenSRE to Slack and hand off DevOps chores for your…
yandex-cloud
Read Yandex Cloud through its REST API. Applies to any question about VMs, metrics, logs, audit events, Kubernetes, managed databases, serverless, networking or any other Yandex Cloud resource. Never shell out to the yc CLI — it is not how this agent reaches Yandex Cloud and is usually not installed.