@passioncode-ai/fabric-agent-adapter 0.83.1 → 0.85.0

package/plugins/fabric-agent-adapter/skills/building-fabric-services/scripts/sample_service.py

@@ -218,7 +218,8 @@ class Service:
             if not fs.feed_data_problems(data):  # data the feed refuses never undoes the answer
                 extra = {"data": data}
                 break
-        self.log.append("board.slot.returned", "info", "%s took the %s slot back%s; the service enrolls once the "
+        # Its own kind (FAA-325): the slot is still away; the kit says board.slot.returned once it is held again.
+        self.log.append("board.slot.taken_back", "info", "%s took the %s slot back%s; the service enrolls once the "
                         "other session lets it go." % (who, capability, how),
                         subject={"type": "capability", "id": capability}, **extra)
         return {"capability": capability, "moved": False}

package/plugins/fabric-agent-adapter/skills/creating-fabric-agents/SKILL.md

@@ -9,7 +9,7 @@ license: AGPL-3.0-only OR LicenseRef-PassionCode-Commercial
 compatibility: Requires filesystem access and Python 3.9+. Exact schema checks additionally need git, Node.js, pnpm, and the pinned fabric-agent-contract checkout (a public repository). Works without those tools in an explicitly degraded structural-check mode. Ships in one plugin with adapting-projects-to-fabric, whose scripts it reuses.
 metadata:
   author: PassionCode.ai
-  version: "0.83.1"
+  version: "0.85.0"
   contract-version: "0.1.0"
   contract-commit: "93e7db164c7829e2783b4b7705c4113bb5201131"
 ---
@@ -70,7 +70,7 @@ board (or, with a grant, its scan tool over the source projects), and build from
 | **Which profile?** | `mcp` when Fabric owns planning and retries and the capability has a schema; `a2a` when the provider owns its task lifecycle end to end; `local-runner` when it is a terminal agent Fabric drives. When unsure, the sibling skill's profile-selection reference is the decision record |
 | **What effects?** | declare anything money-, deletion-, or publication-adjacent now; the host's floor and grants are designed against these declarations, and one discovered late invalidates the admission |
 | **Where does it run, for whom?** | this machine, the person's cloud or an organization's, and in which contexts (`personal`, `org:<id>`): one instance per context, its secrets, spend and journal never crossing into another. Record which capabilities need the person's machine and what they answer when it is off (references/where-agents-run.md) |
-| **Does it run around the clock?** | the uptime category, decided now: **A** when a capability must answer while the owner's machine is off or asleep (a schedule, a chat, callers outside the machine): the agent gets a cloud host from its first release, Heroku unless a reason names another (the person's or an organization's cloud, or PassionCode's for a paid agent), built as an online service (building-fabric-services, the remote placement) and started from the cloud Fabric-server template once that template's repository exists; **B** when nothing needs it then: local only, no cloud (a local service, or a CLI or local runner when nothing calls it while the owner is away), reading an A agent's data through that agent's MCP or API. The README names the category (references/where-agents-run.md) |
+| **Does it run around the clock?** | the uptime category, decided now: **A** when a capability must answer while the owner's machine is off or asleep (a schedule, a chat, callers outside the machine): the agent gets a cloud host from its first release, Heroku unless a reason names another (the person's or an organization's cloud, or PassionCode's for a paid agent), built as an online service (building-fabric-services, the remote placement) and started from the cloud Fabric-server template, whose repository an agent reads from the instance's settings or the estate's agent map, never from this skill; **B** when nothing needs it then: local only, no cloud (a local service, or a CLI or local runner when nothing calls it while the owner is away), reading an A agent's data through that agent's MCP or API. The README names the category (references/where-agents-run.md) |
 | **Whose data does it keep?** | the kind of its data, named in the README: **U** when it keeps data of people outside the owner's team (messages, support requests, reviews, customers' answers), which never enters the source, a copy of it or its tests: the source holds the schema and synthetic fixtures, and a copy starts with an empty store of its own; **O** when it keeps none (a tool or service for other agents, the owner's reports, runs and media). In doubt, U. Either way the data lives in `<agent>-data` (a kit service: the kit's folders; on a cloud host: a durable store), apart from the source, from the first commit (references/source-and-instance.md) |
 | **Who holds the copyright?** | the holder the agent's `LICENSE` names: a new private agent (built for its owner or for a client) gets a `LICENSE` naming that holder in its first commit, all rights reserved unless the owner gives other terms; a public product keeps its own licence (a PassionCode.ai repository: the organization's, step 2) |
 | **Which models?** | per lane (planning, replies, volume work, media): a primary, a fallback of another vendor, an open-vendor model where it reads what users send, the input kinds each reads, named as catalogue models in the profile, narrowed by classes on the spend policy or grant (references/models-and-providers.md) |
@@ -139,7 +139,7 @@ lists them. An agent someone builds for themselves, or a private agent built for
 is its owner's choice and none of those files is required of it, but its first commit carries a `LICENSE` naming
 the copyright holder the grill recorded (all rights reserved unless the owner gives other terms); it covers the
 holder's own files, and the kit files the adapter copies in keep their own licence (AGPL-3.0-only OR
-LicenseRef-PassionCode-Commercial).
+LicenseRef-PassionCode-Commercial), each service kit module in its own SPDX line.
 
 **Every credential the agent uses is read by name, never held as a value**, from its first commit:
 the README's Configure section lists key NAMES. With Project Observatory installed, the agent finds

package/plugins/fabric-agent-adapter/skills/creating-fabric-agents/references/where-agents-run.md

@@ -106,8 +106,8 @@ Every new agent is given an uptime category on its first day, and the README nam
   (the places above), and a paid agent stays in PassionCode's cloud (selling an agent, below). On a host with no
   Fabric hub it answers what reaches its own origin (its schedule, its chat, its API); a call Fabric admits needs a
   cloud hub, as running while the laptop is closed says. The agent is an online service (building-fabric-services, the remote placement), often
-  alongside its local instance, and starts from the cloud Fabric-server template once that template's repository
-  exists. Its data lives where the cloud reaches it, in a durable store (a database, an object store, the private
+  alongside its local instance, and starts from the cloud Fabric-server template, whose repository an agent reads from
+  the instance's settings or the estate's agent map; this public skill names no repository or address for it. Its data lives where the cloud reaches it, in a durable store (a database, an object store, the private
   `<agent>-data` repository), never the host's own folder, which a platform such as Heroku does not keep across a
   restart.
 - **B, local only.** Nothing needs it while the machine sleeps: no cloud, and a local service only when other

package/plugins/fabric-agent-adapter/skills/creating-fabric-agents/scripts/admit.py

@@ -87,34 +87,54 @@ def template_step(folders: List[str]) -> Dict[str, object]:
                    "%d folder(s), %d error(s)" % (len(folders), len(errors)), errors)
 
 
-def _terms_file(named: Optional[str]) -> Path:
+# The one private-terms reader (FAA-328): test/validate.py, the repository's own check, reads the list with these
+# functions too, so the two never drift. They match as org-index's check_private.py does: case-insensitive, not inside a
+# run of ASCII letters and digits, longest term first. Callers report paths and lines only, never a term.
+def terms_file(named: Optional[str] = None) -> Path:
+    """The list: `named`, else `$PASSIONCODE_PRIVATE_TERMS`, else ~/.config/passioncode/private-terms.txt; `~` expanded."""
     named = named or os.environ.get(PRIVATE_TERMS_ENV)
     return Path(named).expanduser() if named else Path.home() / ".config/passioncode/private-terms.txt"
 
 
-def _files(root: Path) -> List[str]:
+def terms_pattern(lines: List[str]) -> "Optional[re.Pattern[str]]":
+    """One alternation of the list's terms (a line that is blank or starts with # is none), longest first; None when
+    the list holds no term."""
+    terms = sorted({t.strip() for t in lines if t.strip() and not t.strip().startswith("#")}, key=len, reverse=True)
+    if not terms:
+        return None
+    return re.compile(r"(?<![A-Za-z0-9])(?:" + "|".join(re.escape(t) for t in terms) + r")(?![A-Za-z0-9])", re.IGNORECASE)
+
+
+def term_lines(text: str, pattern: "re.Pattern[str]") -> List[int]:
+    """The 1-based numbers of the lines of `text` that hold a term."""
+    return [n for n, line in enumerate(text.splitlines(), 1) if pattern.search(line)]
+
+
+def listed_files(root: Path) -> List[str]:
+    """Every tracked file and every untracked one Git does not ignore, as check_private.py reads a work tree; outside a
+    work tree, or where git cannot run, every file but the usual generated folders, never nothing."""
     if (root / ".git").exists():
-        listed = subprocess.run(["git", "-C", str(root), "ls-files", "-z", "--cached", "--others", "--exclude-standard"],
-                                capture_output=True, check=False)
-        if listed.returncode == 0:
+        try:
+            listed = subprocess.run(["git", "-C", str(root), "ls-files", "-z", "--cached", "--others", "--exclude-standard"],
+                                    capture_output=True, check=False)
+        except OSError:
+            listed = None
+        if listed is not None and listed.returncode == 0:
             return sorted({rel for rel in listed.stdout.decode("utf-8", "replace").split("\0") if rel})
     return sorted(p.relative_to(root).as_posix() for p in root.rglob("*")
                   if p.is_file() and not SKIP.intersection(p.relative_to(root).parts))
 
 
 def private_terms_step(root: Path, named: Optional[str]) -> Dict[str, object]:
-    path = _terms_file(named)
+    path = terms_file(named)
     if not path.is_file():
         return _result("private-terms", "NOT_RUN", "no private-terms list on this machine (set %s): not a pass"
                        % PRIVATE_TERMS_ENV)
-    lines = path.read_text(encoding="utf-8").splitlines()
-    terms = sorted({t.strip() for t in lines if t.strip() and not t.strip().startswith("#")}, key=len, reverse=True)
-    if not terms:
+    pattern = terms_pattern(path.read_text(encoding="utf-8").splitlines())
+    if pattern is None:
         return _result("private-terms", "PASS", "the list holds no term")
-    pattern = re.compile(r"(?<![A-Za-z0-9])(?:" + "|".join(re.escape(t) for t in terms) + r")(?![A-Za-z0-9])",
-                         re.IGNORECASE)
     hits: List[str] = []
-    files = _files(root)
+    files = listed_files(root)
     for rel in files:
         try:
             data = (root / rel).read_bytes()
@@ -122,7 +142,7 @@ def private_terms_step(root: Path, named: Optional[str]) -> Dict[str, object]:
             continue
         if b"\0" in data[:8192]:
             continue
-        found = [str(n) for n, line in enumerate(data.decode("utf-8", "replace").splitlines(), 1) if pattern.search(line)]
+        found = [str(n) for n in term_lines(data.decode("utf-8", "replace"), pattern)]
         if found:
             hits.append("%s: line%s %s" % (rel, "s" if len(found) > 1 else "", ", ".join(found)))
     return _result("private-terms", "FAIL" if hits else "PASS",

package/plugins/fabric-agent-adapter/skills/creating-fabric-agents/scripts/check_dashboard_link.py

@@ -46,6 +46,9 @@ def has_jwt(text):
     return False
 
 
+# Kept here, not imported: this gate runs alone in a dashboard's own checkout with no kit beside it, and its three copies
+# stay byte-identical (test_dashboard_links.py). It is broader than `fabric_secrets` on purpose (FAA-136, FAA-187 (4)): a
+# page link that only looks like a secret is refused, since a link is never a place for one.
 # The contract's SECRET_VALUE at the pin (DEC-0151): `Bearer ` at a value's start in any case, or a provider key with its
 # prefix in its canonical case after a boundary (the start, a character that is no ASCII letter or digit, or an escape);
 # an `sk-` key holds a digit and an upper-case letter with 16 or more characters, or a digit with 32 or more.

package/plugins/fabric-agent-adapter/skills/observing-products/SKILL.md

@@ -12,7 +12,7 @@ compatibility: >-
   from Project Observatory's vault by slot name when it is installed.
 metadata:
   author: PassionCode.ai
-  version: "0.83.1"
+  version: "0.85.0"
 ---
 
 # Observing products

package/plugins/fabric-agent-adapter/skills/working-with-fabric/SKILL.md

@@ -13,7 +13,7 @@ compatibility: >-
   this plugin; Codex/Kimi use command hooks; Hermes/OpenCode/OpenClaw/Pi use native context plugins after opt-in.
 metadata:
   author: PassionCode.ai
-  version: "0.83.1"
+  version: "0.85.0"
 ---
 
 # Working with Fabric

package/plugins/fabric-agent-adapter/skills/working-with-fabric/scripts/fabric_session_env.py

@@ -58,7 +58,7 @@ SESSION_ID = re.compile(r"[A-Za-z0-9._:-]{1,128}")
 LOOPBACK_ORIGIN = re.compile(r"http://127\.0\.0\.1:(\d{1,5})")
 
 
-VERSION = "0.83.1"
+VERSION = "0.85.0"
 
 
 HOSTS = {"claude": "Claude Code", "codex": "Codex", "kimi": "Kimi Code",