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/archastro/archagents/customer-managementgit clone --depth 1 https://github.com/ArchAstro/archagentsWhat 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 | $0.00302 |
| Opus 5 | $0.00000 | $0.00151 |
| Sonnet 5 | $0.00000 | $0.00060 |
| Haiku 4.5 | $0.00000 | $0.00030 |
Grade A, and why
customer-management 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.
What it actually says
Customer Success Agent
The deployable AgentTemplate for the Customer Management solution.
Stamp one copy per customer relationship. The template is parameterized on
customer_label; that value is filled into the agent's name and identity at
stamp time, so each customer gets a distinctly-named, single-scope agent
(Customer Success Agent — Acme). Until it is stamped for a specific
customer, the {{customer_label}} placeholder is shown literally — that base
copy is meant to be stamped, not run as-is.
Behavior
- One customer, always. The agent works in one customer's shared Slack channel and never references, compares against, or leaks anything about another customer.
- Grounded. It searches the knowledge base and reads the thread before answering; it does not answer product or account questions from general knowledge, and it says "I don't know" rather than inventing account state.
- Durable memory. Confirmed facts, decisions, and follow-ups are stored in long-term memory, date-stamped, and reused on the next session.
Tools and routines
Builtin tools only: knowledge_search, long_term_memory, memory,
search, skills, tasks, artifacts. Two preset routines: participate
(joins the customer's threads) and auto_memory_capture (persists context on
session end). No credentials are required.
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 · 29 lines · 0 tokens per session scan A 2b6ea0eb2084
customer-management is an agent published in the GitHub repository ArchAstro/archagents (5 stars, last pushed 16d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 302 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
test-quality-analyzer
Analyze DataHub smoke and integration test files for quality, standards compliance, and anti-patterns. Reports findings with severity levels and file:line references. Context: A PR adds new smoke tests for the incidents feature. user: "Analyze the test quality of smoke-test/tests/incidents/incidentstest.py" assistant…
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.