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 eriknewton/sanctuary-framework --skill sanctuarygit clone --depth 1 https://github.com/eriknewton/sanctuary-frameworkWrote 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/eriknewton/sanctuary-framework/sanctuary)<a href="https://agentmods.dev/skills/eriknewton/sanctuary-framework/sanctuary"><img src="https://agentmods.dev/badge/skills/eriknewton/sanctuary-framework/sanctuary/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/eriknewton/sanctuary-framework/sanctuary"><img src="https://agentmods.dev/badge/skills/eriknewton/sanctuary-framework/sanctuary.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 Memory Poisoning · line 133 Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.Fix: Protect agent memory and state from modification by untrusted content. Use read-only memory for critical instructions and validate all state changes.
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 | $0.00204 | $0.03902 |
| Opus 5.5 | $0.00082 | $0.01561 |
| Sonnet 5.5 | $0.00041 | $0.00780 |
| Haiku 4.5 | $0.00020 | $0.00390 |
Grade A, and why
sanctuary 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 today.
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 — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sanctuary Framework
Sanctuary gives your agent protection it can prove: encrypted state, self-custodied identity, selective disclosure, portable reputation, zero-knowledge proofs, federation, posture auditing, and a verifiable security posture. The tools in this skill are the Charter layer, the cooperative MCP surface. The Castle Wall (kernel-level egress enforcement) is armed separately and is live on macOS; on Linux it is proven for Ubuntu 24.04 (x86-64) only; Windows is roadmapped. Every capability claim traces to a row in the Assurance Matrix.
Install, verify, uninstall
npx -y @sanctuary-framework/[email protected] protect --claude-code --no-open --no-dashboard
npx -y @sanctuary-framework/[email protected] doctor --json
npx -y @sanctuary-framework/[email protected] uninstall
-y skips npm's install prompt; without it npx asks before fetching the package in a terminal. --no-dashboard skips the per-call dashboard, which otherwise serves in the foreground until you stop it; wrap still persists the agent record so a separately-running sanctuary dashboard (or a later wrap) sees the harness; start it later with sanctuary dashboard. --no-open runs headless: it skips opening the dashboard in a browser and switches custody confirmation to the non-interactive mode protect already uses whenever stdin is not a terminal.
Substitute --openclaw, --hermes, --cursor, --cline, --mastra, or --wrap <path-to-config> for the harness you run. With protect, the MCP server entries in the config are rewritten to point at Sanctuary. The harness's own built-in tools are not MCP calls and are not touched. The rewrite is tested on representative harness paths and is not proven for every harness version (Assurance Matrix, Agent wrapping). A plain npx -y @sanctuary-framework/[email protected] entry beside other MCP servers starts the server but does not gate calls to those servers. doctor --json is read-only. uninstall keeps keys, state, and the audit log; protect --unwrap restores the original harness configuration. The passphrase stays in the platform credential store: never put a passphrase or recovery key in a harness configuration, never read staged recovery material, and never ask the operator to paste it into chat.
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.
- today Changed 1829f4865dc9
- 5d ago Changed c3d26f9566b8
- 9d ago Changed 32991256224f
- 15d ago Changed d94d3f98c826
- 16d ago Changed · +10 tokens per session 1cc48532c572
- 22d ago Changed · -3 tokens per session 2de01ae93062
- 23d ago Changed · +1 lines 57733c4ae578
- 24d ago Changed · -23 lines · +77 tokens per session 615f8cf96198
- 29d ago Changed 2f2965b29cc8
- 1mo ago First seen · 234 lines · 120 tokens per session scan A 384d1e1860ab
sanctuary is a skill published in the GitHub repository eriknewton/sanctuary-framework (10 stars, last pushed today), licensed Apache-2.0. It adds 204 tokens to every session and 3,902 once invoked, about $0.0008 per session on Opus 5.5. 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.
sanctuary compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| sanctuaryeriknewton/sanctuary-framework | Skill | 10 | today | Apache-2.0 | A | 204 | Claude Code |
| agent-reachPanniantong/Agent-Reach | Skill | 109k | yesterday | MIT | A | 197 | Claude Code, Codex |
| x-signal-agentomnara-ai/omnara | Skill | 3.4k | yesterday | Apache-2.0 | A | 53 | Claude Code, Codex |
| start9-frontendStart9Labs/start-technologies | Skill | 2.0k | yesterday | MIT | A | 133 | Claude Code |
| composiodesplega-ai/agent-swarm | Skill | 1.0k | today | MIT | A | 128 | Claude Code |
| review-offered-taskdesplega-ai/agent-swarm | Skill | 1.0k | today | MIT | A | 33 | Claude Code |
| data_analystsiddsachar/row-bot | Skill | 1.7k | today | Apache-2.0 | A | 26 | Claude Code, Codex |