trust-checks SKILL.md skillB
trust-checks from tykimos/agent-native-agent is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add tykimos/agent-native-agent/trust-checks@git:20261008.3245bd0Pins git:20261008.3245bd0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- B, 16 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- tykimos/agent-native-agent
- Path
skills/trust-checks/SKILL.md- Versions recorded
- 3
- Latest version
- git:20261008.3245bd0, committed Oct 8, 2026
- First version recorded
- Sep 30, 2026
- Size
- 4,900 bytes
- Repository stars
- 128
- License
- AGPL-3.0
Port ANA's System › Trustworthiness · Safety · Security tabs — checks computed on the server from real state (agent running, structured log, tool failure rate, interruptions, plan usage; permission-bypass flags, risky commands the agent actually ran, approval gate; bind address, sign-in, CSRF, token handling, credential file mode, secrets saved in the user's data) shown as ok / info / warn / risk cards with a risk badge. Use when the user says "AI 신뢰성", "안전", "보안", "trust", "위험한 명령", "보안 점검…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# trust-checks — can I rely on this agent, and is it safe?
An ANA agent often runs with permission prompts off and can reach the user's machine and data. These three tabs answer "is it working, what could it break, who can reach it" from **real state**, not static text. Every check must be computed. If you can't compute a check, report `info` ("unknown"); never report `ok` by default.
## Checks (base ANA)
| Tab | Check | Source | Status rule |
|---|---|---|---|
| Trustworthiness | Agent is running | `ctx.hasSession()` + `ctx.agentAlive()` | risk if down |
| | Chat comes from the agent's own log | `agentLog.read(target)` available | warn if screen-mirror only |
| | Tool calls that failed | `isError` over recent tool calls in the log | warn ≥ 10%, risk ≥ 30% |
| | Interrupted turns | `Interrupted` events in the log | info, warn ≥ 5 |
| | Plan usage 5-hour / weekly | `/api/limits` data (Claude) or rollout `rate_limits` (Codex) | warn ≥ 80%, risk ≥ 95% |
| | Waiting on you | pending diff proposals + open requests | info when > 0 |
…Read the whole file at its exact version.
How to install
mdr add tykimos/agent-native-agent/trust-checks@sha256:90b6bd7a0941178cThe line at the top of this page pins the label git:20261008.3245bd0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_2llclszf5yzy7fyc)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Sep 30, 2026 and Oct 8, 2026, about one every 4 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261008.3245bd0 latest | 2026-10-08 | 3245bd0 | 4,900 B | B | view · diff |
| git:20261006.fce69a0 | 2026-10-06 | fce69a0 | 4,884 B | B | view · diff |
| git:20260930.d354f65 | 2026-09-30 | d354f65 | 4,792 B | B | view |
Audit of the latest version
- fail: No curl or wget piped into a shell (matched: curl \| sh)
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (4900 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.
Source
tykimos/agent-native-agent · 128 stars · license AGPL-3.0 · pushed 2026-10-15 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_2llclszf5yzy7fyc GET https://markdownregistry.com/api/v1/resolve?ref=tykimos/agent-native-agent/trust-checks GET https://markdownregistry.com/api/v1/blob/90b6bd7a0941178c337a4591c0e3f3f06c066ab1595800c5d3899304c8ec7183
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.