diff --git a/AGENTS.md b/AGENTS.md
index 11a04e0..b957cff 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -19,14 +19,7 @@ it: a GitHub Actions composite (`action.yml`) and a GitLab CI Catalog component
 ## Architecture
 
 A human at a shell, the GitHub Action, and the GitLab CI component all invoke the same
-`semvertag tag`, and `semvertag/` is short enough to read. What reading it will not tell you is why
-four of its shapes are load-bearing rather than incidental: the two forge adapters are independent
-copies sharing only the RFC 8288 pagination loop
-([ADR-0001](docs/adr/0001-forge-providers-not-unified.md)) and their status-to-error ladders stay
-duplicated for the same reason ([ADR-0003](docs/adr/0003-error-translators-not-tabled.md)); there is
-one verb and no `doctor` preflight ([ADR-0006](docs/adr/0006-no-doctor-preflight-command.md)); and
-the composite action deliberately does not check the repository out
-([ADR-0007](docs/adr/0007-composite-action-does-not-check-out.md)).
+`semvertag tag`, and `semvertag/` is short enough to read.
 
 ## Cutting a release (maintainers)
 
@@ -39,41 +32,8 @@ Release and move `v0` by hand, or cut a new patch tag.
 
 ## Workflow
 
-**The spec for a change is its PR body**, not a committed file: why, design, non-goals, verification,
-reviewed with the diff. There is no change file and no lane to choose. A trivial PR (typo, dep bump,
-formatter, CI tweak) ships a conventional-commit title with no body ceremony.
-
-Two things outlive the PR, and there are exactly two places to put them: an alternative **rejected**
-with reasoning becomes an ADR in [`docs/adr/`](docs/adr/) (`NNNN-slug.md`, sequential, with a revisit
-trigger), and real work **not scheduled** becomes a GitHub issue. There is no third state, and no
-separate truth-home directory — a behaviour change is reviewed with the diff, not promoted to a page.
-
-### Where a fact goes
-
-Four homes, one owner each:
-
-| Home | Holds |
-|---|---|
-| `semvertag/` | anything readable from the source — the default |
-| a named test | an **invariant**: must stay true, and a change could silently break it |
-| `docs/adr/` | a rejected alternative, with the reasoning that would otherwise be re-litigated |
-| `README.md` and `docs/` | anything a user needs |
-
-Before writing a line anywhere:
-
-> Can an agent get this by reading `semvertag/`? → **don't write it.**
-> Would a wrong change here fail a test? → it belongs **in the test**, not in prose.
-> Does a user need it? → **`README.md` / `docs/`**.
-> Otherwise it does not get written.
-
-**Prose about mechanism has no home. There is no file to add a paragraph to.** This file included:
-it is always loaded, so a line that restates a docstring, a justfile comment, or `pyproject.toml`
-costs every turn and rots in two places at once. The workflow files here carry dense explanatory
-comments; re-telling them in this file is the failure mode to watch for.
+Real work **not scheduled** becomes a GitHub issue.
 
 An invariant is a test whose name is the claim, with a docstring opening `INVARIANT:` and a second
 paragraph naming **what breaks it** — design rationale, not a report of what this one test catches.
-Nothing enforces that docstring shape; it is read at review time. A relative link to an ADR *is*
-checked — CI runs lychee `--offline` over every `.md` — but a path named in a docstring or a comment
-is not. Both ADRs and `INVARIANT:` docstrings ratchet: nothing prunes a record once its call is
-settled. Keeping them lean is a standing habit.
+Nothing enforces that docstring shape; it is read at review time.