@brainervirus/workit-core 8.11.0 → 8.12.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@brainervirus/workit-core",
-  "version": "8.11.0",
+  "version": "8.12.0",
   "private": false,
   "description": "Workit shared core — task, policy, evidence, review, decision, worker, and writer state for agentic coding workflows",
   "keywords": [

package/skills/workit-fanout/references/brief.md

@@ -25,7 +25,7 @@ REPORT: branch, head SHA, files changed, each VERIFY command with its exit code,
   each ACCEPTANCE line met / not met, rulings you made (`workit ledger ruling`),
   anything out of scope as a follow-up, not a diff.
 STANDING: <the standing orders, verbatim: `workit ledger standing add` records each one>
-  export WORKIT_SESSION_ID=<lead>-w<n>   (set by the lead for a worker's commits; a verifier brief sets none: the host derives it)
+  <identity line, per the lead's host: see "Identity and git rules per host" below>
 ```
 
 ## Worked example
@@ -97,17 +97,32 @@ is the slice's STUCK threshold in `workit fanout status`. `"fanIn":
 }
 ```
 
+## Identity and git rules per host
+
+`workit fanout brief` renders the identity line and RULES 1 and 3 for the
+lead's host (`WORKIT_HOST`, or the host's own session variable):
+
+| Lead's host | Identity line | RULES 1 and 3 |
+| --- | --- | --- |
+| Claude Code | `Session: your env already has WORKIT_SESSION_ID (the lead's id)`; no `export`, no hand-minted id (SubagentStart issues `<lead>:<agent id>` only to verifiers and reviewers, never to an implementer, so the worker's commits carry the lead's id) | plain git: `git fetch origin && git switch -c <branch> origin/<base>` and `git commit -m <msg> --trailer "Workit-Session: <the env WORKIT_SESSION_ID>" --trailer "<host attribution>" -- <paths in SCOPE>`, because worktree isolation refuses every `workit git` form |
+| OpenCode, Pi, Codex | `Session: your host already sets WORKIT_SESSION_ID in your shell`; export nothing (the host derives the id and an export cannot replace it) | `workit git branch <branch> --base <base>`, `workit git commit -m <msg> -- <paths in SCOPE>` |
+| Cursor, CI, a plain terminal | `export WORKIT_SESSION_ID=<lead>-w-<slice>` (`+try<n>` for a race attempt): declared, not checked | the same `workit git` forms |
+
+`MODE: resume` is `git switch <branch>` on every host (the branch exists; the
+lead removed the dead worker's worktree after its exit, so the switch
+succeeds), then continue from its head. A race attempt (`--attempt <n>`)
+follows its host's row with the branch `<branch>-try<n>`.
+
 ## Worker rules (`workit fanout brief` appends them as RULES)
 
-1. First command. `MODE: new`: `workit git branch <branch> --base <base>` (the
-   worktree may start on a name that breaks branch policy). `MODE: resume`:
-   `git switch <branch>` (the branch exists; the lead removed the dead
-   worker's worktree after its exit, so the switch succeeds), then continue
-   from its head.
+1. First command: create the branch (`MODE: new`; the worktree may start on a
+   name that breaks branch policy) or `git switch` to it (`MODE: resume`),
+   as the table above says for your host.
 2. Decide ambiguities yourself and record them:
    `workit ledger ruling "<what>" --why "<why>" --cost-if-wrong "<cost>"`.
    Stop only for an irreversible action, a security-sensitive one, or a side
    effect outside the worktree.
-3. Commit with `workit git commit -m "<msg>" -- <paths in SCOPE>`; push only if
-   the brief says so.
+3. Commit as the table says (`workit git commit -m "<msg>" -- <paths in SCOPE>`,
+   or plain git with the trailers on Claude Code); push only if the brief says
+   so.
 4. Never record a verdict on your own work.

package/skills/workit-fanout/references/orchestration.md

@@ -73,8 +73,8 @@ Which session id a verdict carries, per host:
 - Claude Code: subagents inherit your `WORKIT_SESSION_ID`, so the
   SubagentStart hook names each `verifier` and `reviewer` its own session
   (`<lead>:<agent id>`), passed exactly as `--session`: the hook records the
-  ids it issues, and only those are accepted. An `implementer` uses the
-  session its brief exports for its workit commands.
+  ids it issues, and only those are accepted. An `implementer` is issued
+  none: it keeps your id in its env for its workit commands and commits.
 - Codex: the `workit-verifier` and `workit-reviewer` agents get the same
   issued `--session` from SubagentStart; any other verifier passes
   `--as verifier`.
@@ -84,8 +84,13 @@ Which session id a verdict carries, per host:
 - Cursor (no session id in the shell): the brief sets the verifier's
   `WORKIT_SESSION_ID=<lead>:verifier-<random hex>`, a fresh suffix per
   verifier. Cursor cannot check it: it is declared, not derived.
-- A worker's commits use `export WORKIT_SESSION_ID=<lead>-w-<slice>` from
-  `workit fanout brief`. A verdict under an id you made up (`<lead>-v1`,
+- A worker's commits: on Claude Code the lead's `WORKIT_SESSION_ID` already in
+  its env (SubagentStart issues no implementer id; plain git, no export, so the
+  lead authored those commits: a verifier or reviewer records the verdict, not
+  the lead); on OpenCode, Pi and Codex the id the host already
+  puts in its shell; on Cursor, CI and a plain terminal the brief's
+  `export WORKIT_SESSION_ID=<lead>-w-<slice>` (see `brief.md`, "Identity and
+  git rules per host"). A verdict under an id you made up (`<lead>-v1`,
   `<lead>:v1`) is refused as `forged_session` on Claude Code, Codex,
   OpenCode and Pi.
 

package/src/check-config.ts

@@ -47,6 +47,8 @@ export type CheckConfig = {
   gates: Partial<Record<"testing" | "verification", string[]>>;
   /** Set when the config exists but cannot be used; gates then fail closed. */
   error: string | null;
+  /** The raw `resources` block of workit.checks.json (heavy-check limits; core/resources.ts reads it). */
+  resources?: unknown;
 };
 
 const isRecord = (value: unknown): value is Record<string, unknown> =>
@@ -228,7 +230,7 @@ export function loadCheckConfig(cwd: string): CheckConfig {
       gates[gate as (typeof GATES)[number]] = names as string[];
     }
   }
-  return { source: CHECKS_FILE, root, checks, gates, error: null };
+  return { source: CHECKS_FILE, root, checks, gates, error: null, resources: parsed.resources };
 }
 
 const sameArgv = (left: readonly string[], right: readonly string[]): boolean =>

package/src/checks.ts

@@ -169,6 +169,13 @@ export type RunOptions = {
   onStderr?: (text: string) => void;
   /** For tests: the platform whose spawn rules apply (default: process.platform). */
   platform?: NodeJS.Platform;
+  /**
+   * POSIX only: argv that runs the command (e.g. a `systemd-run … --` scope
+   * prefix). A --shell command then runs as `<wrapper> /bin/sh -c <cmd>`.
+   */
+  wrapper?: readonly string[];
+  /** Called with the child's pid once it started (on POSIX also its process group id). */
+  onSpawn?: (pid: number) => void;
 };
 
 // cmd.exe metacharacters (the cross-spawn set).
@@ -272,7 +279,12 @@ export function runCheckCommand(argv: readonly string[], options: RunOptions): P
   const buffer = new TailBuffer(MAX_LOG_BYTES);
   const redactor = new StreamRedactor((text) => buffer.push(text));
   const env = options.env ?? process.env;
-  const plan = planSpawn(argv, { platform, env, cwd: options.cwd, shell: options.shell });
+  const wrapped = Boolean(options.wrapper?.length) && platform !== "win32";
+  const command = wrapped
+    ? [...options.wrapper!, ...(options.shell ? ["/bin/sh", "-c", argv[0]] : argv)]
+    : argv;
+  const shell = wrapped ? false : options.shell;
+  const plan = planSpawn(command, { platform, env, cwd: options.cwd, shell });
   return new Promise((resolve) => {
     let settled = false;
     let timedOut = false;
@@ -334,6 +346,7 @@ export function runCheckCommand(argv: readonly string[], options: RunOptions): P
       finish(null, null, (error as NodeJS.ErrnoException).code ?? (error as Error).message);
       return;
     }
+    if (child.pid) options.onSpawn?.(child.pid);
     if (!windows)
       for (const name of FORWARDED_SIGNALS) {
         const handler = () => {

package/src/core/methods.ts

@@ -105,9 +105,9 @@ Autonomy contract:
   ship), go to the effective endpoint in \`workit grant show\`: \`pr\` opens
   the PR; \`green\` babysits it without asking (workit-ship) to merge-ready,
   never merging; \`merged\` also lands it with \`workit pr merge\`.
-  Merge and release need a workspace grant, which only the user raises from
-  their own terminal (a blocked verb names how); never edit workspaces.json or
-  route around a grant_required.
+  Merge and release need a workspace grant only the user raises from their
+  own terminal (a blocked verb names how); never edit workspaces.json or route
+  around a grant_required.
 - Ask only for a product or preference choice, or for authority you lack, with
   your recommended answer. Facts are yours: read, run or prototype.
 - Label claims measured (run and seen this session), inferred, or guess.
@@ -116,17 +116,19 @@ Autonomy contract:
 - Author is not verifier: your own verdict is \`--self\`, shown as
   self-reviewed, never verified. High risk, \`verification: independent\` and
   merge need a verifier with its own host-derived session (a subagent's
-  \`<lead>:<child>\`, set by the host); never hand-mint one such as \`<lead>-v1\`.
+  \`<lead>:<child>\`); never hand-mint one (\`<lead>-v1\`).
 - A local-commit endpoint does not imply PR readiness, and a local commit alone
   is not evidence of a requested remote push. Before reporting delivery,
   reconcile every requested item and observe it (\`workit verify-delivery\`).
 
 CLI first: if a step has one right answer, use the \`workit\` verb instead of
 hand-running git, gh or glab (\`workit git|pr|ci|check\`; \`workit help\` lists
-the rest; without \`workit\` on PATH, \`npx -y @brainervirus/workit-cli\`).
+the rest; lacking \`workit\`, \`npx -y @brainervirus/workit-cli\`).
+Add Co-Authored-By via \`workit git commit --trailer "K: v" -- <paths>\`, never
+an amend.
 The task families (task, policy, evidence, finding, decision, worker, state)
 are optional continuity for tracked work; each branch has one implicit task,
-created by the first recording, and a solo edit needs no task. A decision
+and a solo edit needs no task. A decision
 records a user choice; it never authorizes an effect. Judge work once with
 \`workit policy assess --judge risk= behavior= product-choice= plan=\`.
 Across repositories, bind each item to its checkout, branch and endpoint,