lkn
/
ollie-core
Archived
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
654 Commits 1 Branch 0 Tags 17 MiB
Commit Graph

8 Commits

Author SHA1 Message Date
Levi Neely 2ea13cd611 restructure: drop pkg/, split tools/execute into focused packages 
- Drop pkg/ prefix (Go anti-pattern): ollie/pkg/X → ollie/X
- Split tools/execute god package:
  - execute/: shell execution, sandboxing, elevation client, remote SSH
  - tools/: interfaces + registry + discovery + schema parsing
  - detach/: background process management (ring buffer, signal)
- Promote internal/sandbox → sandbox/
- Absorb config/ into agent/config.go (agent definition loading)
- Merge remote/ into execute/remote.go (RemoteServer)

All tests pass.
 2026-07-29 18:10:25 +02:00
Levi Neely c5b16e739e Restructure: pkg/core interface, internal/ packages, cmd/ollie TUI entry point 
- pkg/core: public Core interface (Submit/Prompt/Interrupt) + Event/EventHandler types
- internal/agentcore: AgentCore implementation of Core (session, tools, commands)
- internal/tui: TUI package (readline loop, splitInput, signals, bracketed paste)
- internal/agent,backend,config,exec,mcp,tools: moved from root to internal/
- cmd/ollie: minimal main() that wires agentcore + tui together (~130 lines)
- mkfile: update build target to ./cmd/ollie

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-09 21:00:04 +02:00
Levi Neely 1c4c3049ac agent: reduce repetition and blind writes with dedup caches and range tracking 
- Revised system prompt with concrete prohibitions against verbosity and
  premature stopping
- Added GenerationParams (max_tokens, temperature, frequency/presence penalty)
  threading from agent config through backend ChatStream calls
- Added stall detection: emits "stalled" role on max-steps hit or zero tool
  calls with content, surfaces as "stalled" in status bar
- Added per-session file read range tracking: warns on overlapping re-reads,
  blocks file_write unless the target range was previously read
- Added general tool-call dedup: warns on exact (name, args) repeats for
  non-file tools
- Both caches invalidated on /compact and /clear; file read cache invalidated
  per-path on file_write
- Updated all agent configs with documenting defaults for new generation params

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
 2026-04-06 21:29:20 +02:00
Levi Neely ab575ee385 backend,agent,main: retry on HTTP 429 with live countdown 
Return RateLimitError from openai backend on HTTP 429, parsing the
Retry-After header (integer seconds or HTTP-date). The agent loop retries
up to 3 times with exponential backoff (5s/10s/20s) when no header is
given, emitting per-second countdown ticks. The status bar renders
"retry {N}s" during the wait.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-04 11:25:05 +02:00
Levi Neely 474eb84e2d agent,backend: drop non-streaming path 
Both backends already implement streaming; the non-streaming fallback
was dead code that added complexity.

- Backend interface now requires only ChatStream; the separate
  StreamingBackend interface and Chat method are removed
- Both OpenAIBackend and OllamaBackend lose their Chat methods and the
  stream=bool parameter on their internal doChat helpers
- Loop.Run is simplified: one streaming path, no streamed flag, no
  skippedCalls map, no shouldStop helper, no runStreamStep indirection
- "call" events are now emitted solely in the act phase, not in the
  stream phase, so dedup tracking is no longer needed

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-03 20:11:10 +02:00
Levi Neely d60a1b9215 Add streaming support for bot output and tool execution 
- backend: StreamingBackend interface + StreamEvent type
- backend/ollama: ChatStream with line-by-line JSON-LD decoder
- backend/openai: ChatStream with SSE delta parser + tool arg accumulation
- agent/loop: runStreamStep reads deltas, emits OutputMsg per chunk,
  deduplicates call/tool display between stream and outer loop
- main: agentMsg tea type streams incremental updates to the TUI;
  plain string buf replaces strings.Builder (avoids copy-by-value panic);
  drainAgent cancels in-flight goroutines cleanly on user interrupt
 2026-04-01 12:52:58 +02:00
Levi Neely f9ba6c8bce Track and display token usage per request 
Add Usage{InputTokens, OutputTokens} to backend.Response. Both
OllamaBackend (prompt_eval_count/eval_count) and OpenAIBackend
(prompt_tokens/completion_tokens) populate it. The loop emits a
'usage' OutputMsg after each Chat call; the UI displays it as
[↑N ↓N tokens].

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-01 00:31:03 +02:00
Levi Neely 127f428620 Add backend package: common LLM interface for ollama and OpenAI-compatible APIs 
Defines a Backend interface with shared Message/Tool/ToolCall types.
OllamaBackend wraps /api/chat; OpenAIBackend covers OpenAI, OpenRouter,
and any compatible API. Factory reads OLLIE_BACKEND, OLLIE_API_URL,
OLLIE_API_KEY.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-03-31 22:40:42 +02:00