kami
968cbfa973
Found while live-QAing freestyle_planning on a 12B local model: - list_dir tool: recursive, .gitignore-aware listing so weak models stop flooding context with `ls -R` over node_modules/build/dist. Wired into the fileRead toggle + advertised to the planner (architect_freestyle). - ContextClassifier: assistantToolCall turns are STRUCTURED, so the token pruner never shreds the model's own tool-call history — that was causing amnesia loops (re-issuing calls it had already made). - Retire instruction-doc LLMLingua pruning (DOC_SOURCE_TYPES emptied): it fused load-bearing procedural text into unparseable soup. The static block stays small by dropping CLAUDE.md at the loader instead. - AgentInstructionsLoader: load only AGENTS.md, not CLAUDE.md — the latter targets the outer assistant and polluted the agent's stage context. - DefaultSessionReducer: WorkflowFailed flips session status to FAILED (was stuck ACTIVE forever, so clients/approval loops never saw a terminal). - ShellTool: run shell command lines (cd/&&/pipes) via `sh -c`; unrunnable program is recoverable instead of an uncaught IOException killing the stage; malformed argv (non-string/collapsed-array) rejected with guidance. - llmlingua sidecar: cap force_tokens to max_force_token (big docs blew the assert and 500'd, so doc pruning silently failed open). Tests added/updated across all of the above.
40 lines
1.3 KiB
TOML
40 lines
1.3 KiB
TOML
id = "freestyle_planning"
|
|
start = "analyst"
|
|
|
|
# analyst writes no files, but it owns task framing: task_search/task_context (read-only) find
|
|
# existing work; task_create (T2, approval-gated — a task is an event-log entry, not a file write)
|
|
# opens a single task; task_decompose (T2, one approval for the whole graph) splits a goal with
|
|
# dependency seams into parent + DEPENDS_ON-linked children. Either way the analysis names the task
|
|
# id the run will work, so the architect threads it into the plan's implementation stages.
|
|
[[stages]]
|
|
id = "analyst"
|
|
prompt = "prompts/analyst_freestyle.md"
|
|
produces = [{ name = "analysis", kind = "analysis" }]
|
|
allowed_tools = ["file_read", "list_dir", "shell", "task_search", "task_context", "task_create", "task_decompose"]
|
|
token_budget = 16384
|
|
max_retries = 2
|
|
|
|
[[stages]]
|
|
id = "architect"
|
|
requires_approval = true
|
|
inject_artifact_kinds = true
|
|
prompt = "prompts/architect_freestyle.md"
|
|
needs = ["analysis"]
|
|
produces = [{ name = "execution_plan", kind = "execution_plan" }]
|
|
token_budget = 16384
|
|
max_retries = 2
|
|
|
|
[[transitions]]
|
|
id = "analyst-to-architect"
|
|
from = "analyst"
|
|
to = "architect"
|
|
condition_type = "artifact_validated"
|
|
condition_artifact_id = "analysis"
|
|
|
|
[[transitions]]
|
|
id = "architect-to-done"
|
|
from = "architect"
|
|
to = "done"
|
|
condition_type = "artifact_validated"
|
|
condition_artifact_id = "execution_plan"
|