# `AshGrant.ExprStringify`
[🔗](https://github.com/jhlee111/ash_grant/blob/v0.22.0/lib/ash_grant/expr_stringify.ex#L1)

Converts `Ash.Expr` terms and scope filter values into human-readable,
LLM-friendly strings.

Ash's default `Inspect` for expressions is already quite readable, but it
uses internal reference tuples like `{:_actor, :id}` or `:_tenant`.
Downstream consumers (the `AshGrant.Explanation` JSON surface, future
Phoenix dashboard, and Ash AI tools) need strings that mirror the DSL
syntax users write in their resources — `^actor(:id)`, `^tenant()`,
`^context(:key)`.

This module is **read-only and best-effort**: the output is a display
string, never parsed back. Unknown terms fall back to `inspect/1` and the
function never raises.

## Examples

    iex> AshGrant.ExprStringify.to_string(true)
    "true"

    iex> expr = AshGrant.Info.resolve_scope_filter(MyApp.Post, :own, %{})
    iex> AshGrant.ExprStringify.to_string(expr)
    "author_id == ^actor(:id)"

## Public API contract

This module is part of AshGrant's public introspection surface consumed by
`ash_grant_phoenix` and `ash_grant_ai`. Breaking changes are tracked in
CHANGELOG.

# `to_string`

```elixir
@spec to_string(term()) :: String.t()
```

Converts a scope filter value to a human-readable string.

Accepts:
- `true` / `false` / `nil`
- `Ash.Expr` terms (any struct inspected via Ash's protocol)
- Arbitrary Elixir terms (falls back to `inspect/1`)

---

*Consult [api-reference.md](api-reference.md) for complete listing*
