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 ericrisco/rsc-harness --skill agent-safetygit clone --depth 1 https://github.com/ericrisco/rsc-harnessWrote 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/ericrisco/rsc-harness/agent-safety)<a href="https://agentmods.dev/skills/ericrisco/rsc-harness/agent-safety"><img src="https://agentmods.dev/badge/skills/ericrisco/rsc-harness/agent-safety/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/ericrisco/rsc-harness/agent-safety"><img src="https://agentmods.dev/badge/skills/ericrisco/rsc-harness/agent-safety.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 58 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Privilege Escalation · line 64 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- medium Excessive Agency · line 34 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00079 | $0.02319 |
| Opus 5 | $0.00039 | $0.01159 |
| Sonnet 5 | $0.00016 | $0.00464 |
| Haiku 4.5 | $0.00008 | $0.00232 |
Grade A, and why
agent-safety scanned grade A with 1 finding 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 9d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
return subprocess.run(cmd, shell=True, capture_output=True, text=True).stdout How it starts
The opening of the file, as written. The whole thing — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent safety
You are the security review for an agent's agency, not for its code. The loop works,
tools are wired, memory persists — your job is to make that autonomy bounded. If you
want to review ordinary endpoints, auth, or secrets handling, that is
../secure-coding/SKILL.md — this skill is the Agentic Top 10, the risks that exist only
because a model has tools and autonomy. If the loop or tools do not exist yet, that is
../building-agents/SKILL.md. You arrive after both.
references/threat-model.md carries the OWASP Agentic Top 10 2026 risks mapped to the
controls below, the pre-ship guardrail checklist, and the incident-triage flow for "the
agent did X" — open it when you are reviewing before ship or reconstructing an incident.
The ownership split
Agent security splits into four layers — Model · Harness · Tools · Environment. The model provider owns only the Model layer (alignment, refusals). Everything else is yours: the Harness (loop, memory, context assembly), the Tools (what the agent can do), and the Environment (creds, network, blast radius). Do not outsource a layer you own to "the model is aligned."
Three excesses cause almost every agentic incident. Cut all three:
- Excessive functionality — tools the task never needs.
- Excessive permissions — broader scopes/creds than the tool needs.
- Excessive autonomy — acting without checking back when it should.
The operating principle is least agency: autonomy is earned per task, not defaulted.
Scope limits
- Declare the allowed task domain as a hard boundary in the system prompt. Why: an undeclared scope is an infinite scope; "you are a refund assistant; you do not touch payroll" is a constraint a reviewer can check.
- Deny by default — the agent starts with zero tools. Each tool earns its place by a task justification. Why: an opt-out tool list grows; an opt-in list stays minimal.
- Segregate the instruction channel from the data channel. System/developer prompt = trusted instructions. Everything the agent reads at runtime = data, never instructions. Why: this single boundary is what stops indirect injection (LLM01).
What ships with it
3 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.
- 9d ago First seen · 170 lines · 79 tokens per session scan A b1e010bc563f
agent-safety is a skill published in the GitHub repository ericrisco/rsc-harness (74 stars, last pushed yesterday), licensed MIT. It adds 79 tokens to every session and 2,319 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
securing-agentic-ai-tool-invocation
Apply least-privilege tool allowlisting, identity binding, and human-in-the-loop controls for agent tool calls.
orchestrating-llm-attacks-with-pyrit
Build multi-turn, Crescendo, and Tree-of-Attacks-with-Pruning (TAP) automated attack chains against conversational LLM agents using Microsoft PyRIT, with adversarial chat and scorer feedback loops.
defending-llms-with-guardrails
Deploy Llama Guard, NeMo Guardrails, and LLM Guard input/output scanners as runtime defenses.
red-teaming-llms-with-garak
Run NVIDIA garak probe suites against an LLM endpoint to test for jailbreaks, prompt injection, data leakage, and toxic generation, then interpret the hit-rate report for triage and reporting.
testing-prompt-injection-in-rag-pipelines
Probe RAG applications for prompt injection via poisoned retrieved context and embedding manipulation.
merge-aliases
Folds two surface names for the same backend system into one canonical entity, keeping every original mention individually retrievable, and refuses to merge pairs that only share spelling.