Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only Effect-TS coding guide with disclosed, purpose-relevant examples for HTTP, AI providers, subprocesses, uploads, and persistence.
Installers should treat this as a developer reference skill. Use it when you want Effect-TS guidance, and review generated code before running it, especially examples involving subprocesses, external AI/HTTP providers, database mutations, uploads, or durable/background execution.
The description says to use the skill when 'writing, debugging, or reviewing Effect code' and whenever code imports from 'effect' or any '@effect/*' package. This is a wide trigger scope without negative examples or clearer constraints, which could cause the skill to activate for many ordinary coding conversations or repositories that merely contain those imports.
The manifest describes this skill as guidance for writing, debugging, or reviewing Effect code across areas like errors, concurrency, services, streams, and schema. This file is specifically a concurrency reference, but it also documents process-spawning APIs that execute external commands, which is a materially different operational capability than concurrency primitives alone.
The section documents effect/unstable/process and shows invoking node, git, and head through a process spawner. For a skill presented as an Effect-TS coding guide, especially within a concurrency reference, teaching shell-command execution is an extra capability with security implications that is not clearly justified by the stated purpose.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
// GET request
const response = yield* client.execute(
HttpClientRequest.get("https://api.example.com/users")
)
// POST with JSON body
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
// GET request
const response = yield* client.execute(
HttpClientRequest.get("https://api.example.com/users")
)
// POST with JSON body
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
// GET request
const response = yield* client.execute(
HttpClientRequest.get("https://api.example.com/users")
)
// POST with JSON body
The multipart upload guidance explicitly states that uploaded files are persisted to disk and exposed via a filesystem path, but it does not warn about the security and operational implications of writing untrusted user content to local storage. In a developer guide, this omission can lead adopters to deploy unsafe defaults, causing disk exhaustion, unsafe retention of sensitive files, or downstream processing of attacker-controlled files without validation.
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
## Other API Removals & Renames
| v3 | v4 |
|------------------------------------------|-----------------------------------------------------------------------------------|
| `Layer.scoped(Tag, eff)` | `Layer.effect(Tag, eff)` — strips `Scope` from requirements automatically |
| `Effect.async((resume) => ...)` | `Effect.callback((resume, signal) => ...)` — `signal: AbortSignal` is positional |
Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.
| `Schedule.exponential("100 millis", 1.5)` | Custom growth factor |
| `Schedule.fibonacci("100 millis")` | Fibonacci backoff |
| `Schedule.fixed("5 seconds")` | Fixed interval (accounts for elapsed time) |
| `Schedule.forever` | Repeat indefinitely |
| `Schedule.once` (v3 only) | Run once more — v4: use `Schedule.recurs(0)` |
| `Schedule.jittered` | Add randomness (combine with other schedules) |
| `Schedule.take(n)` | Bound any schedule to n iterations (v4 idiom) |
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation source, and
configuration files.
"Object" form shall mean any form resulting from mechanical transformation or
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation source, and
configuration files.
"Object" form shall mean any form resulting from mechanical transformation or
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation source, and
configuration files.
"Object" form shall mean any form resulting from mechanical transformation or
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation source, and
configuration files.
"Object" form shall mean any form resulting from mechanical transformation or
This markdown file includes example code that sends prompts to external AI providers and later documents tracing of model-call metadata, but the skill description does not warn that user-supplied content may be transmitted to third-party services or captured in observability spans. For markdown files, SQP-2 applies when behaviors affecting privacy are described without disclosure.
Detected: suspicious.exposed_secret_literal