Commit Graph

9 Commits

Author SHA1 Message Date
435da49263 Prefer JSON-bearing envelope fields, skip metadata, in Claude CLI unwrap
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
The first CUST-WP-0045 canary retry after 9de0f49 still failed schema
validation with `Expecting value: line 1 column 1 (char 0)`. The original
allowlist returned envelope.result verbatim, which on longer prompts
carries the model's conversational preamble ("Triage report generated
and returned via structured output. Key signals: ..."), not the
schema-enforced JSON. The actual structured payload lives in a different
envelope field whose name varies across CLI versions.

Make the unwrap order-aware:
  1. Scan envelope fields and return the first one whose value parses as
     JSON (dict, list, or a string that loads cleanly). Skip well-known
     metadata keys (type, usage, total_cost_usd, etc.) so telemetry can
     never be mistaken for the model payload.
  2. Fall back to the original text-field allowlist only when no field
     carries JSON, so non-schema callers via this same code path still
     see the model's prose.
  3. Surface the raw envelope as last resort.

This is robust against unknown envelope shapes — as long as the schema-
enforced JSON appears somewhere in a non-metadata field, the adapter
will find it.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-02 12:44:25 +02:00
9de0f495db Pass --output-format json with --json-schema and unwrap CLI envelope
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
The Claude Code adapter previously passed --json-schema alone. On Claude
CLI 2.1.160 that combination still emits the model's conversational
preamble on stdout while the schema-enforced structured payload ships on
a sidecar channel the adapter cannot read. Result: callers requesting
structured output got prose that fails JSON parsing downstream — exactly
the failure mode the activity-core CUST-WP-0045 daily triage canary hit
on 2026-06-01 ("Triage report generated and returned via structured
output. Key signals:..." → json.loads error at column 1).

Fix: when --json-schema is set, also pass --output-format json. The CLI
then writes a JSON envelope on stdout. The adapter unwraps it by
probing a small allowlist of known text-bearing fields (result,
result_text, content, text, output). Unknown envelope shapes fall
through to raw stdout so the operator can introspect the structure and
extend the allowlist.

The unwrap path is only triggered when --json-schema was set, so non-
schema callers keep the existing raw-stdout behavior.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-02 10:20:24 +02:00
b12d1af8bb Support Claude Code JSON schema execution
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
2026-05-21 03:19:27 +02:00
82e3c07928 Preserve llm-connect run config in server mode
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
2026-05-19 20:55:02 +02:00
c11c6afa3f Implement-LLM-WP-0005-cost-model-estimators
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
2026-05-19 05:02:20 +02:00
a27945101c Adaptive routing initial version
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
2026-05-18 11:38:12 +02:00
c4ad4bb9f2 Add adaptive cost-quality routing primitives
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
2026-05-17 21:32:27 +02:00
d51d6303e2 feat: WP-0003 — RoutingPolicy (FR-2) and HTTP serve mode (FR-1)
Some checks failed
CI / test (3.10) (push) Has been cancelled
CI / test (3.11) (push) Has been cancelled
CI / test (3.12) (push) Has been cancelled
FR-2 RoutingPolicy:
- RoutingPolicy + RoutingRule dataclasses in llm_connect/routing.py
- resolve(task_type, estimated_cost_per_1k=None) with cost-cap fallback
- Exported from llm_connect.__init__; contract doc at contracts/functional/routing-policy.md
- 11 tests covering rule match, cost-cap, fallback, unknown type, no-match

FR-1 HTTP serve mode:
- LLMServer in llm_connect/server.py (stdlib http.server, zero extra deps)
- POST /execute + GET /health; CLI via python -m llm_connect.server
- [server] optional-dep group added to pyproject.toml
- Contract doc at contracts/functional/server.md
- 9 tests: health, round-trip, 400/404/500 errors, config forwarding
- Added "mock" provider to factory for CLI default

All 101 tests pass.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-01 22:34:00 +00:00
d71f4114d1 feat: WP-0001 foundation + WP-0002 core extensions
WP-0001 — Foundation & GAAF Baseline
- SCOPE.md, ARCHITECTURE-LAYERS.md, contracts/ tree
- .claude/rules/ stubs filled (architecture, stack, boundary)
- 57 tests (pytest), pyproject.toml with ruff+mypy, CI workflow

WP-0002 — Core Extensions (FR-4 + FR-3)
- FR-4: BudgetTracker (thread-safe) + LLMBudgetExceededError +
  optional RunConfig.budget_tracker + enforcement in all adapters
- FR-3: async_execute_prompt on LLMAdapter ABC (asyncio.to_thread
  fallback) + native asyncio.create_subprocess_exec in ClaudeCodeAdapter

81 tests passing.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-01 22:24:14 +00:00