| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
Teach provider config and auth resolution about the Codex Responses
dialect, including user-facing `style = "openai_responses"` with
`dialect = "codex"`. Serialize and parse Responses traffic with
provider-specific reasoning replay, assistant phase metadata, and robust
function-call assembly keyed by `output_index` so streamed tool inputs
survive proxy quirks and empty terminal payloads.
Also persist thinking origins and message metadata across sessions, add
the Anthropic interleaved-thinking header switch, write per-session
debug logs, and improve the TUI and scripts for inspecting tool output
and session costs.
|
|
Add provider.mergeHeaders and use it in both the openai_chat and
anthropic_messages providers so config extra_headers ride on the model
request, merged onto the built-in header set.
|
|
|
|
|
|
Conversation.addUserMessage now takes a []ContentBlock (symmetric with
addAssistantMessage). Introduce a thin addUserText wrapper in agent.zig
for the plain-text case and update every call site in agent.zig and
anthropic_messages_json.zig accordingly.
|
|
|
|
A facade that forwards every method 1:1 is worse than paring the real type
down to the intended surface and aliasing it. conversation.Conversation's
interface is already the public surface we want (init/deinit, the
add*/replace* builders, and messages/allocator as plain data fields), so
public.zig now aliases it straight through and Agent.conversation() returns
a borrowed *Conversation for in-place surgery. Drops the ConversationData
alias and the pointer-wrapper.
Merged addAssistantMessageWithUsage into addAssistantMessage(blocks, ?usage)
on the real type (separate method deleted), so the alias has the intended
one-method shape; all call sites updated.
Only Agent and Stream remain true facades -- they have genuinely-internal
fields plus API-shaping renames (Stream.phase->state, Phase->State) an alias
can't express.
|
|
converted the main agent loop from push-based (into callbacks on a
Receiver vtable) to pull-based, where `next()` re-enters a state-machine
Stream until the next event can be returned.
|
|
|
|
|
|
|
|
Replace all references to the "system" layer with "base" to better reflect
its role as the foundational extension/tool layer in panto's hierarchy.
The layer hierarchy now consistently uses: project > user > base.
Includes:
- Update agent README and build.zig documentation
- Refactor tool registry and config module to support layered lookups
- Add TestHarness abstraction for cleaner test setup
- Improve JSON serialization with wire-encoded tool names
- Add glob pattern matching for tool/extension discovery
|
|
|
|
OpenAI and Anthropic have tool details (id, name) converge on different
timelines. In all cases they are available in `onBlockComplete` which
comes with the full block, and previously we had an optional `BlockMeta`
argument to `onBlockStart` - the anthropic timeline.
We removed the `BlockMeta` from `onBlockStart` since it was always going
to be unreliable, and now have an explicit `onToolDetails` instead. This
allows us to provide the tool id and name as early as possible from the
openai provider (could be in the middle of content deltas), and provide
a reliable signal across providers.
Also, fix test output.
|
|
Close each tool_use on next-index delta (and at finalize) so receivers
see one start/delta/complete trio per tool, matching Anthropic. Log+drop
fragments for already-closed indices.
Remove BlockMeta from the Receiver interface: tool id/name can't be
reliably known at onBlockStart under OpenAI's streaming model. Receivers
get identity from the assembled ContentBlock at onBlockComplete. CLI
renders 'tool: {args} : (name)' accordingly.
|
|
|
|
|
