> For the complete documentation index, see [llms.txt](https://docs.skillplus.xyz/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.skillplus.xyz/use-cases.md).
# Use Cases
SkillPlus can be used by individuals, teams, authors, and platforms that need a trust layer for AI skills.
## Use-case map
```mermaid
flowchart TD
A[SkillPlus trust signal]
A --> B[Individual developers]
A --> C[Security and platform teams]
A --> D[Skill authors]
A --> E[Marketplaces]
A --> F[Enterprise governance]
B --> B1[Check before installing]
C --> C1[Approve before internal use]
C --> C2[Respond to supply-chain incidents]
D --> D1[Publish badges and fix findings]
E --> E1[Rank and display safer skills]
F --> F1[Control what agents can use]
```
## Before installing a skill
A developer finds a useful skill on GitHub or skills.sh.
Before installing it, they submit the URL to SkillPlus and review the report.
This helps answer:
* Does the skill reference suspicious domains?
* Does it ask the agent to run unsafe commands or read sensitive files?
* Does its dependency chain include any confirmed-compromised package?
* Did the multi-agent audit see suspicious intent — and did adversarial verification confirm it?
## Supply-chain incident response
This is where the persistent graph changes the game. A real example:
In March 2026, `litellm` 1.82.7/1.82.8 on PyPI were poisoned with credential-harvesting malware (MAL-2026-2144). Several published AI skills instruct agents to `pip install litellm` unpinned.
```mermaid
flowchart LR
A[Advisory lands] --> B[SkillPlus blacklist entry
litellm 1.82.7–1.82.8]
B --> C[Graph propagation
no re-scan needed]
C --> D[Every dependent skill flips to High
report · badge · homepage · search]
D --> E[Reports name the poisoned versions
+ pin-to-safe-version recommendation]
```
* **For a security team**: query "which of our approved skills depend on litellm?" and get an immediate, evidence-backed answer — while snapshot scanners still say *Pass* because their verdicts predate the advisory.
* **For a platform**: badges embedded in listings flip automatically; users see the risk at the point of install.
* **For a skill author**: the report tells them exactly what to do — pin to a version published after the cleanup.
## Team security review
A team wants to approve third-party skills before employees use them.
SkillPlus gives the team a repeatable review process:
```mermaid
flowchart LR
A[Submit skill] --> B[Review report]
B --> C[Check rating, evidence, supply chain]
C --> D{Decision}
D --> E[Approve]
D --> F[Reject]
D --> G[Request changes
using the report's recommendations]
```
Because ratings are live, an approval is not a one-time event: if a dependency of an approved skill is later compromised, the skill's rating flips to High and the team sees it on their next check — or through the SDK in their dashboard.
## Skill author workflow
An author publishes a skill and wants users to trust it.
1. Scan the skill and review the findings.
2. Apply the report's recommendations (pin dependency versions, declare `allowed-tools`, remove risky patterns).
3. Re-scan — the drift-aware pipeline produces a fresh report for the new content.
4. Embed the badge in the README:
```md
```
## Marketplace and registry integration
A marketplace wants installation pages to carry a security signal.
Using the SDK, each listing can show the SkillPlus rating, a badge, and a link to the full report — with deduplicated scan-if-missing behavior for skills that have not been scanned yet. Popularity ranks discovery; SkillPlus ranks trust.
## CI publishing gate
A team publishing internal skills adds SkillPlus to CI:
```mermaid
flowchart LR
A[Skill repo PR] --> B[CI: skillplus.query]
B --> C{rating}
C -- safe --> D[Publish]
C -- medium --> E[Require reviewer sign-off]
C -- high --> F[Block + attach report recommendations]
```
## Enterprise agent governance
An organization controls which skills its agents may load. SkillPlus provides the policy signal:
* allowlist skills rated safe, with whitelist overrides for explicitly accepted risks
* block anything with an active supply-chain hit — enforced by the same code path that floors the rating
* audit trail: every decision links to an evidence-backed report pinned to the exact content version reviewed
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.skillplus.xyz/use-cases.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.