jamesmoss/agent-school
Investigates a problem area in the codebase and finds or creates a tessl tile (rules, docs, skills) to teach agents how to handle it correctly. Use when agents keep making the same mistakes around a library, design pattern, or convention.
56
70%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
W011: Third-party content exposure detected (indirect prompt injection risk)
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
Third-party content exposure detected (high risk: 0.80). The SKILL.md workflow (Phase 2) explicitly instructs the agent to run `tessl search` against the public Tessl registry and to evaluate/install tiles from that registry, meaning the agent will fetch and interpret third-party (potentially untrusted) tile content that can change tool use and steer future agent behavior.
Where we found it
Tessl registry (via tessl search)
domain · 4 sites
The plugin instructs the agent to search the public Tessl registry and evaluate third-party tile content (summaries, rules, docs, skills), which is untrusted external content that could steer agent behavior.
SKILL.md
29
## Phase 2: Search the Tessl Registry
SKILL.md
41
Run **multiple searches** using the `tessl search` MCP tool to cast a wide net.
SKILL.md
16
either an existing one from the registry or a new one
SKILL.md
111
**Always search the registry first** — don't reinvent the wheel if a well-maintained tile already exists
W012: Unverifiable external dependency detected (runtime URL that controls agent)
What this means
The skill fetches instructions or code from an external URL at runtime, and the fetched content directly controls the agent’s prompts or executes code. This dynamic dependency allows the external source to modify the agent’s behavior without any changes to the skill itself.
Why it was flagged
Potentially malicious external URL detected (high risk: 0.80). The SKILL instructs the operator to run "tessl install tessleng/tile-creator" (and similar `tessl install` calls like `tessl install jamesmoss/agent-school`), which fetches remote tile code from the registry at runtime that will be loaded as steering rules/skills and therefore can directly control agent prompts/instructions.
Where we found it
tessleng/tile-creator
dependency · 3 sites
The plugin explicitly instructs the agent to install the tessleng/tile-creator tile from the remote Tessl registry at runtime, fetching external code that becomes steering rules/skills controlling agent behavior.
SKILL.md
83
Install the `tessleng/tile-creator` tile which provides a dedicated skill for authoring new tiles:
SKILL.md
86
tessl install tessleng/tile-creator
SKILL.md
101
The `tessleng/tile-creator` skill is purpose-built for this
tessl install <workspace>/<tile-name>@<version>
command · 2 sites
The plugin instructs the agent to install arbitrary tiles from the Tessl registry based on search results, fetching remote code that becomes steering rules/skills at runtime.
SKILL.md
67
If the user wants to install an existing tile, use the `tessl install` MCP tool:
SKILL.md
70
tessl install <workspace>/<tile-name>@<version>
- Audited
- Security analysis