auditor-lambda 0.3.19 → 0.3.21

package/skills/audit-code/audit-code.prompt.md

@@ -1,46 +1,17 @@
 ---
-description: Autonomous local loop code auditing - advances deterministic audit state, delegates bounded review tasks, and ingests validated results
+description: Autonomous local loop code auditing - loads one backend-rendered audit step at a time
 argument-hint: [target-dir]
 allowed-tools: [Read, Bash, Glob, Grep, Agent]
 ---
 
-# `/audit-code` Execution Directive
+# `/audit-code` Loader
 
 You are the audit-code orchestrator for this conversation. The user-facing
-surface is only `/audit-code`; do not ask the user to choose backend commands,
-providers, models, paths, or batching strategy during normal operation.
+surface is `/audit-code`, but the backend owns every audit workflow branch.
 
-Your job is to advance the deterministic state machine, delegate bounded
-semantic review when the host supports subagents, and let the backend validate
-and ingest results mechanically.
+## Loader
 
-## Core Guardrails
-
-- Do not edit source files during semantic review. The deterministic
-  `auto_fixes_applied` executor may run formatter/remediation commands before
-  review; that is part of the backend workflow.
-- Do not manually merge audit results, manually update coverage, or manually
-  edit audit state.
-- Do not read result schemas or completed result payloads into context unless
-  a backend command fails and the error explicitly requires diagnosis.
-- Do not inspect individual subagent result files after dispatch. Validation
-  and ingestion are backend responsibilities.
-- CRITICAL: Do not use your `Read` tool to read `entry.prompt_path` or JSON schemas into your own context window. The subagent will read them. Pass the path literally.
-- Prefer subagent dispatch for semantic review whenever the host exposes an
-  Agent/subagent tool.
-- Treat the user's `/audit-code` request as explicit authorization to launch
-  review subagents in parallel. Do not ask for a separate delegation request
-  before using available Agent/subagent tools.
-- Do not use `browser_subagent` for semantic review of source code unless the
-  task explicitly requires browser-based validation.
-- If the host cannot dispatch subagents, complete exactly one assigned review
-  task, run the provided ingestion command, then stop. The user can run
-  `/audit-code` again to continue from fresh context.
-
-## Step 1 - Advance Deterministic State
-
-First, make sure the repository has the minimal local assets required by the
-current host:
+First, make sure the repository has current local audit assets:
 
 ```bash
 audit-code ensure --quiet
@@ -52,115 +23,24 @@ Inside the `auditor-lambda` repository itself, use:
 node audit-code.mjs ensure --quiet
 ```
 
-Then run:
+Then ask the backend for exactly one next step:
 
 ```bash
-audit-code
+audit-code next-step
 ```
 
 Inside the `auditor-lambda` repository itself, use:
 
 ```bash
-node audit-code.mjs
+node audit-code.mjs next-step
 ```
 
-Parse only the command JSON envelope needed for routing:
-
-- `audit_state.status`
-- `handoff.active_review_run.run_id`
-- `handoff.artifacts_dir`
-- `handoff.active_review_run.task_path`
-- `handoff.active_review_run.prompt_path`
-- `handoff.active_review_run.pending_audit_tasks_path`
-- `handoff.active_review_run.audit_results_path`
-- `handoff.active_review_run.worker_command`
-
-If status is `"active"`, deterministic progress was made. Run Step 1 again.
-
-If status is `"complete"`, skip to Step 5.
-
-If status is `"blocked"` and the blocker is not semantic review, report the
-blocker verbatim and stop.
-
-If status is `"blocked"` for semantic review, continue to Step 2.
-
-## Step 2 - Dispatch Review Work
-
-When the host supports subagents, prepare a dispatch plan by default:
-
-```bash
-audit-code prepare-dispatch --run-id <run_id> --artifacts-dir <artifacts_dir>
-```
-
-Read only `<artifacts_dir>/runs/<run_id>/dispatch-plan.json`.
-
-In a single message, launch one Agent/subagent call per dispatch-plan entry:
-
-```text
-Agent({ description: entry.description, prompt: "Read and follow the audit instructions in: " + entry.prompt_path })
-```
-
-Do NOT use your `Read` tool to load `entry.prompt_path` into your context window. The subagent has its own context window and will read the file.
-
-If the host supports per-subagent model selection, use `entry.model_hint.tier`
-as a provider-neutral routing hint (`small`, `standard`, or `deep`). Map it to
-available host models without asking the user to choose model names. If model
-selection is unavailable, ignore the hint and dispatch normally.
-
-If the host supports per-subagent tool restrictions, give review subagents no
-Write tool and allow shell access only for the `audit-code submit-packet`
-command printed in their prompt.
-
-All subagent calls should be launched together. Wait for them to finish.
-
-Subagents own bounded semantic review. They must read only their prompt and
-assigned files, produce the requested `AuditResult[]`, pipe it to the
-`submit-packet` command in their prompt, retry up to 3 times if submission
-fails, and stop. The backend command validates and writes the packet-owned
-result artifacts. They must not use direct file writes, edit source files,
-remediate findings, create extra task results, run unrelated audits, or write
-the worker `result.json` control envelope.
-
-Then run:
-
-```bash
-audit-code merge-and-ingest --run-id <run_id> --artifacts-dir <artifacts_dir>
-```
-
-If `merge-and-ingest` exits non-zero, stop immediately and report the exact
-error. Do not improvise manual merging or state edits.
-
-Loop back to Step 1.
-
-## Step 3 - Single-Task Fallback
-
-Use this path only when the host cannot dispatch subagents.
-
-Read the current review prompt named by `handoff.active_review_run.prompt_path`
-or `.audit-artifacts/dispatch/current-prompt.md`, plus the matching task file
-needed to find `audit_results_path` and `worker_command`.
-
-Complete exactly one assigned review task. If a batch file lists multiple tasks,
-choose the first pending task only. Read only that task's assigned files. Write
-one valid `AuditResult` object, wrapped in a JSON array, to `audit_results_path`.
-
-Run the exact `worker_command` from the task file. Then stop and summarize that
-one bounded step. Do not loop into another semantic review task in the same
-conversation turn.
-
-## Step 4 - Backend Failure Handling
-
-If `prepare-dispatch`, `merge-and-ingest`, or `worker_command` fails:
-
-- stop immediately
-- report the exact command and error output
-- do not manually create prompts, split tasks, merge results, edit state, or
-  remediate application code
-
-Invalid or missing subagent output is a blocker. It should not be silently
-merged or treated as automatic progress.
+Read the returned JSON only far enough to find `prompt_path`, then read and
+follow only that prompt. Do not read packet prompts, schemas, command catalogs,
+or handoff files unless the current step prompt explicitly instructs you to do
+so.
 
-## Step 5 - Present Results
+When a step prompt tells you to continue, run `audit-code next-step` again and
+follow only the newly returned `prompt_path`.
 
-When `audit_state.status` is `"complete"`, do not run the orchestrator again.
-Read `audit-report.md` and present the completed audit with work blocks first.
+Stop when the current step prompt tells you to stop.