Abridge Hello WorldSAFE
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Overview
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
4f83675ca38aOBSERVED · 2026-10-08Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned |
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: abridge-hello-world description: "Run a bounded Abridge smoke test through the documented user workflow using an approved test encounter and no invented API. Use when validating a new tenant or pilot cohort. Trigger with \"smoke test Abridge\"." argument-hint: "[environment] [test-encounter-id]" allowed-tools: Read, Glob, Grep, WebFetch, Write, Edit version: 1.5.0 author: Jeremy Longshore <[email protected]> license: MIT tags: - saas - abridge - smoke-test - consent - note-review model: inherit effort: medium compatibility: Designed for Claude Code; live work requires an authorized Abridge tenant, approved test data, and health-system change authority --- # Abridge Consent-to-Note Pilot Smoke Test ## Overview Prove the smallest useful path: authorized user, designated test patient or synthetic scenario, consent handling, recording, note creation, Web Editor review, Linked Evidence check, and approved final disposition. ## Prerequisites - The authorized Abridge environment, clinical owner, and health-system policy set - Current tenant-specific implementation evidence for every private interface in scope - Synthetic data or the organization's formally approved test-record procedure ## Tool Discipline Use `Read`, `Glob`, and `Grep` to inspect repository configuration, adapters, tests, policies, and existing evidence. Use `WebFetch` only for current official Abridge, HHS, or named EHR documentation. Use `Write` or `Edit` only after confirming scope, environment, owners, patient-data boundary, and approval state. These tools do not confer access to Abridge, an EHR, or a clinical record; return exact operator steps or an approval-gated handoff for live actions. ## Current Contract - Abridge's documented entry path uses its clinician application and Web Editor rather than a public create-session REST tutorial. - Recording guidance requires following the organization's consent policy. - The clinician reviews and edits the generated draft before sending or otherwise finalizing it. ## Authentication Use only the health system's provisioned Abridge application access, SSO, administrative role, or tenant-specific partner authentication documented for the approved environment. Do not infer public API credentials, reuse production secrets in tests, or expose tokens and session material. Verify identity owner, least privilege, environment binding, storage, rotation, and revocation before any authenticated action. ## Instructions 1. Confirm the environment, authorized user, approved test-record procedure, consent script, and expected note destination. 2. Use `Read`, `Glob`, and `Grep` to inspect the local test plan, tenant labels, and expected configuration without opening credentials. 3. Run the documented application workflow with synthetic dialogue or the health system's designated non-production encounter. 4. Verify note creation, visible sections, editability, and a sample of Linked Evidence; record outcomes without clinical text. 5. Exercise one safe failure path such as interrupted capture or withheld send, then confirm no unintended chart update. 6. Use `Write` or `Edit` to record the receipt; use `WebFetch` only for current official Abridge instructions. ## Approval Boundaries Do not use a real patient or send content to a production chart unless the health system's controlled test protocol explicitly authorizes it. ## Output Return environment, test identity class, consent path, workflow stages, evidence sample result, chart-write count, failure-path result, and go/no-go. Separate verified facts, tenant-specific evidence, assumptions, and actions still awaiting approval. ## Error Handling | Condition | Response | |---|---| | No designated test encounter exists | Stop and obtain one through the EHR test-data process. | | Patient identity does not match | Stop before recording. | | Draft cannot be reviewed | Do not send or finalize it. | ## Example The example is a redacted operational receipt, not patient data or proof of vendor certification. ```text environment=nonprod; encounter=designated-test; consent=scripted; draft=reviewable; evidence=sample-pass; chart-writes=0; result=pass ``` ## Resources - [Official documentation map](references/official-docs.md) — dated public evidence and the limits of what those sources establish. Read the source map before changing a workflow. Recheck tenant-specific implementation evidence for every interface or capability that public documentation does not define. ## Next Steps Revalidate the evidence date and tenant-specific authority before repeating this workflow in another environment, cohort, care setting, or integration mode.
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__abridge-hello-world.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 4f83675ca38a | SAFE | B | 89 | first audit |
Questions
What does the Abridge Hello World skill do?
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Is Abridge Hello World safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Abridge Hello World access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does Abridge Hello World work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (4f83675ca38a), read on 2026-10-08. The repository is watched, and a new audit runs when it changes — this is the first audit.