The comments claimed a result above the ceiling has no document, which is not
what the code does: it has one, and the dropped tail is what that document
cannot serve.
A tool result now stays whole in its message (up to a storage ceiling) and
is trimmed on the request copy by the request-prep pass, where soft compact
already runs. The trimmed copy names the document that holds the rest, so
every surface (chat-completions, Responses, Messages) shares one rule.
New [agent] hard_max_tool_result_chars bounds the stored body; the wire cap
stays [agent] max_tool_result_chars.
Co-authored-by: CommandCodeBot <noreply@commandcode.ai>
The answer tables lose the third button (two answers, or the prompt's own),
the popup's close and a body click both withdraw it, and the desktop section
says why every popup rings the terminal: the human is at another window by
construction — the announcement waits out the grace period — and no desktop
lets a program raise one.
The body table gains the workspace (last line, `~` for home, following a
rebind), the desktop section says which popups expire and why, and "Go to
terminal" is described as what it does: withdraw, ring, leave the prompt
waiting — no window is raised, because no desktop allows it.
The body is no longer "the reason or nothing": it is the question, the
reason behind `include_reason`, the object of the prompt and the deadline,
with the desktop keeping the lines apart and the terminal flattening them.
The tables gain the answers a question offers (a reply field and the way
back), the channels' click-to-terminal behaviour, and the secret prompt that
is announced nowhere at all.
The table of what a notification answers was out of date: the limit prompt and
the batch confirm now offer their own answers, and a skill write distinguishes
"once" from "this chat". The desktop section now says what actually keeps the
notification up (no expiry, urgency at the default) instead of naming `critical`,
and the follow-up rule — announced at once, past the grace period — is written
where a reader looks for why a prompt appeared without waiting.
The page said a notification answer is treated like a typed one, without saying
that the prompt has to be listening before the notification exists — the grace
period is why. One paragraph says it, together with the price: a question that
offers a reply field reads in polling mode, so no Tab completion for that
prompt.
Uploading a picture and referencing it by id is the one image path the API
offers that we do not use, and the vendor shapes do not line up: the upload is
the same multipart `POST …/files` with a `file` part everywhere, but the object
that comes back is `object`/`bytes`/unix time on OpenAI, `type`/`size_bytes`/
RFC 3339 on Anthropic, and DeepSeek serves both dialects under two base URLs;
the reference block differs a third time per surface, and OpenAI's chat
completions cannot name an image by id at all.
Recorded next to the inline-path limits, with the three constraints that decide
any future wiring: an id belongs to the provider that minted it (like
`provider_blocks`), the OpenAI path implies the Responses convention, and the
token bill is unchanged — uploading only buys the inline size cap, a smaller
request body, and not re-sending the same bytes every round. The Anthropic
per-file limit is marked unverified: its API reference is unreachable from this
host's region.
`image_in` is not just the mention switch any more: the table names where each
image-carrying path is gated, and says why the gate exists — a text-only model
answers `[Unsupported Image]`, and a gateway that cannot read image blocks feeds
the model raw base64, which is easy to mistake for a delivered picture.
A new page, `doc/trash.md`, covers what a delete does now: where a path lands on
each platform, what is still deleted outright (an explicit `permanent=true`, a
host that cleared `[trash]`, scratch paths under the OS temp dir, the
`copy_path` / `move_path` overwrite, the temporary-workspace cleanup), what
happens when a trashing is refused — the file stays and the error says how to
delete it on purpose — and the two things the trash cannot help with: a shell
`rm` and anything a plugin or MCP server does on its own.
`[trash]` joins the example config and the README snippet, the README gains the
one-paragraph version, `AGENTS.md` the tool/config notes (including the confirm
wording a delete now uses), and the default tool playbook tells the model that
`delete_path` is undoable and what `permanent=true` costs.
A prompt that waits on the human is announced after a grace period, and the
notification can carry the answer: allow / deny / review buttons for a
confirmation, a reply field for an `ask_user_line`. `doc/notifications.md` lays
out the two channels (the optional `desktop-notifier` extra, and OSC 777 / OSC 9
written to `/dev/tty` with no dependency), what each prompt kind offers, why the
reason line stays out by default, and the platform limits — no buttons where a
server does not render them, a reply field only under KDE, kitty left with the
bell.
The example config and the README config block gain the `[notify]` section, and
AGENTS.md documents the three prompt hooks and the rule that only the outermost
prompt of a nested read is announced.
Session YOLO skips the soft confirms of `YOLO`-tagged tools — interpreter runs included — while untagged tools keep prompting and a command that escalates privileges keeps its own confirm through `DangerReason(yolo_skip=False)`. A plugin now also sees that answer shape where the danger classifier is documented, and the round's aggregated prompt leaves the covered calls out.
`doc/images.md` holds the message shape, how `@path` attaches a file, the
per-surface mapping, the provider limits, what a live probe against DeepSeek
confirmed (`tool_result` and `function_call_output` images work, `detail: low`
is 3.4x cheaper, an assistant-message image is a 400), and why the Files API is
not used.
AGENTS.md gets the invariants and the mention/config keys; `doc/architecture.md`
and the example TOML point at it.
doc/turns-and-rounds.md defines the two nested units the loop is built on: a round is one request→response plus the tool batch it asked for, a turn is one user message through the explicit stop (0..N rounds), and a session is a stored sequence of turns. It records what bounds each unit (max_rounds, truncation continues, retries), which work is round-scoped (request assembly, context reduction, the aggregated soft confirm) and which state resets per turn (read-before-edit, yyy auto-continue, /yolo once, the todo touched flag, usage totals) versus what outlives it (rows, todo stack, grants, reminder bands).
Linked from the architecture doc, with the short version in AGENTS.md.
The default catalog and the sample config still named `deepseek-v4-flash`,
which DeepSeek keeps only as a compatibility alias. Spell out the mapping
Claude-* ids get server-side (`deepseek-v4-pro` / `deepseek-flash`) instead
of the `deepseek-v4-*` glob the flash rename made stale.
README and the example config cover the [skills] section, the five tools, and
the read/write split; AGENTS.md gains the skills/ package, the tool family, the
config fields, and the /skills + /status surface; architecture.md lists the
package. doc/skills.md is the author guide: layout, the frontmatter subset and
how a broken one degrades, roots and shadowing, what the agent sees, the
grant story for reads and writes, and the warning that a skill is data —
third-party skill directories are untrusted input.
Remove system strategy (legacy maps to developer). synthetic_tool still
forges the call but uses real stack text as the tool result, and now
yields ToolCall/Result events so the CLI flushes and does not glue nag
text to the next assistant stream.
Unset/empty url still maps to durable user-data chat.db.
Explicit :memory: is not rewritten; CLI warns when not quiet.
DatabaseConfig.url default is None; engine treats None as memory.
NCBM