Update portable Tree Ring skill for v0.14

.github/workflows/validate.yml

@@ -4,91 +4,14 @@ on:
   push:
   pull_request:
 
+permissions:
+  contents: read
+
 jobs:
   validate:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@fbc6f3992d24b796d5a048ff273f7fcc4a7b6c09 # v5
 
-      - name: Validate SKILL.md
-        run: |
-          python3 - <<'PY'
-          from pathlib import Path
-          import re
-
-          text = Path("SKILL.md").read_text(encoding="utf-8")
-          readme = Path("README.md").read_text(encoding="utf-8")
-          if not text.startswith("---\n"):
-              raise SystemExit("SKILL.md must start with YAML frontmatter")
-
-          _, frontmatter, body = text.split("---", 2)
-          required = ["name:", "description:", "version:", "tags:", "triggers:"]
-          missing = [field for field in required if field not in frontmatter]
-          if missing:
-              raise SystemExit(f"Missing frontmatter fields: {', '.join(missing)}")
-
-          version_match = re.search(
-              r"^version:\s*(\d+)\.(\d+)\.(\d+)\s*$",
-              frontmatter,
-              flags=re.MULTILINE,
-          )
-          if not version_match:
-              raise SystemExit("SKILL.md must declare a semantic version")
-          version = tuple(int(part) for part in version_match.groups())
-          if version < (0, 13, 0):
-              raise SystemExit("Multi-agent guidance requires skill version >= 0.13.0")
-
-          if "# Tree Ring Memory" not in body:
-              raise SystemExit("Missing Tree Ring Memory heading")
-
-          required_frontmatter = [
-              '"multi-agent"',
-              '"fan-out"',
-              '"fan-in"',
-              '"idempotency"',
-              '"coordination"',
-              '- "multi-agent memory"',
-          ]
-          missing_frontmatter = [
-              marker for marker in required_frontmatter if marker not in frontmatter
-          ]
-          if missing_frontmatter:
-              raise SystemExit(
-                  "Missing v0.13 tags/triggers: " + ", ".join(missing_frontmatter)
-              )
-
-          required_guidance = {
-              "operation-id idempotency": r"--operation-id",
-              "coordinator capability": r"TREE_RING_COORDINATOR_TOKEN",
-              "protected shared writes":
-                  r"project, global, workflow, session, or other shared/non-agent writes",
-              "protected lifecycle writes":
-                  r"ring changes and supersede/delete/redact lifecycle operations",
-              "protected maintenance": r"maintenance with apply or repair flags",
-              "same-host/local-filesystem boundary":
-                  r"one host using a local\s+filesystem",
-              "schema-v3 old-writer fence":
-                  r"schema v3 fences\s+memory inserts, updates, and deletes from old writers",
-              "mixed-version rejection":
-                  r"all mixed-version\s+operation is unsupported",
-          }
-          missing_guidance = [
-              label
-              for label, pattern in required_guidance.items()
-              if re.search(pattern, body) is None
-          ]
-          if missing_guidance:
-              raise SystemExit(
-                  "Missing v0.13 guidance: " + ", ".join(missing_guidance)
-              )
-
-          if "0.13.0 or newer" not in readme:
-              raise SystemExit("README.md must require Tree Ring CLI 0.13.0 or newer")
-
-          blocked = ["OPENAI_API_KEY", "ANTHROPIC_API_KEY", "sk-", "BEGIN PRIVATE KEY"]
-          for marker in blocked:
-              if marker in text:
-                  raise SystemExit(f"Potential secret marker found: {marker}")
-
-          print("SKILL.md validation passed")
-          PY
+      - name: Validate SKILL.md and publication contracts
+        run: python3 scripts/validate.py

PRIVACY.md

@@ -0,0 +1,17 @@
+# Tree Ring Memory Skill Privacy Notice
+
+Effective August 23, 2026
+
+This repository contains agent instructions only. It does not operate a hosted
+service, collect analytics, send telemetry, or include a remote MCP server.
+
+When an agent runs the separately installed Tree Ring Memory CLI, the CLI stores
+the memory content the user chooses in a local SQLite database under the
+configured Tree Ring root. The project does not receive that database or its
+contents. Data leaves the local environment only when the user or another tool
+explicitly exports, syncs, publishes, or otherwise transmits it.
+
+The skill instructs agents to avoid transcripts, credentials, secrets, private
+keys, raw chain-of-thought, and unnecessary sensitive personal data. These
+safeguards do not replace the privacy and data-use terms of the AI host,
+operating system, source-control provider, or another tool the user invokes.

README.md

@@ -36,11 +36,18 @@ brew install tree-ring
 Or use the canonical project install guide:
 <https://github.com/TerminallyLazy/Tree-Ring-Memory#install>
 
-The multi-agent, Coordinated-policy, and schema-v3 guidance in this skill
-requires Tree Ring Memory CLI **0.13.0 or newer**. Before a v0.13 process opens
-an existing store, stop all Tree Ring processes, checkpoint and back up the
-database, and upgrade every CLI, plugin, and bundled worker. Do not use v0.12
-against an upgraded schema-v3 root; all mixed-version operation is unsupported.
+The receipt-backed harness, multi-agent, Coordinated-policy, and schema-v3
+guidance in this skill requires Tree Ring Memory CLI **0.14.0 or newer**. Before
+a current process opens a pre-v0.13 store, stop all Tree Ring processes,
+checkpoint and back up the database, and upgrade every CLI, plugin, and bundled
+worker. Do not use v0.12 against an upgraded schema-v3 root; all mixed-version
+operation is unsupported.
+
+If the CLI is absent or older, the skill reports the limitation. It does not
+install or upgrade software, edit shell configuration, or claim that recall or
+another memory action ran without explicit user permission and observed command
+output. On a host without local shell access, it remains useful as
+memory-lifecycle guidance.
 
 ## What It Teaches
 
@@ -51,6 +58,8 @@ against an upgraded schema-v3 root; all mixed-version operation is unsupported.
 - Use rings deliberately: cambium, outer, inner, heartwood, scar, and seed.
 - Prefer evidence records for evaluated outcomes.
 - Treat source documents as authoritative when memory and source files disagree.
+- Read the applicable DOX-style `AGENTS.md` chain before editing, preview DOX
+  sync output, and never let recalled summaries override or rewrite contracts.
 - Redact, delete, or supersede stale or unsafe memory.
 - Give same-host fan-out workers unique agent profiles and logical operation
   IDs while sharing workflow and attempt-level session IDs.
@@ -68,14 +77,26 @@ against an upgraded schema-v3 root; all mixed-version operation is unsupported.
 - Keep the shared-root claim bounded to concurrent processes on one host and a
   local filesystem; use per-host stores and explicit evidence-preserving fan-in
   across hosts.
+- Distinguish project-local harness configuration from activation, which
+  requires a fresh matching receipt from scoped recall and safe context
+  injection in a new session.
+- Report exact non-active states such as `configured-awaiting-proof`,
+  `needs-trust`, `needs-plugin`, `needs-project-mount`, and `needs-user-review`
+  without modifying trust or manufacturing receipts.
+- Distinguish installed-CLI harness and recall-quality evidence from the full
+  repository-only `scripts/certify-tree-ring.sh` release suite.
 
 ## Source Project
 
 - Framework: <https://github.com/TerminallyLazy/Tree-Ring-Memory>
+- v0.14 release: <https://github.com/TerminallyLazy/Tree-Ring-Memory/releases/tag/v0.14.0>
 - Claude plugin wrapper:
   <https://github.com/TerminallyLazy/tree-ring-memory-claude-plugin>
 - Skill file: [`SKILL.md`](SKILL.md)
 
 ## License
 
 MIT. See [`LICENSE`](LICENSE).
+
+See [`PRIVACY.md`](PRIVACY.md), [`TERMS.md`](TERMS.md), and
+[`SECURITY.md`](SECURITY.md) for data handling, use terms, and disclosures.

SKILL.md

@@ -1,8 +1,9 @@
 ---
 name: tree-ring-memory
 description: Guides AI agents in using Tree Ring Memory for durable recall, project decisions, user preferences, warnings, future seeds, privacy-safe memory capture, and lifecycle-aware forgetting.
-version: 0.13.0
-tags: ["memory", "agents", "recall", "privacy", "projects", "dox", "revolve", "skills", "cli", "multi-agent", "fan-out", "fan-in", "idempotency", "coordination"]
+version: 0.14.0
+license: MIT
+tags: ["memory", "agents", "recall", "privacy", "projects", "dox", "revolve", "skills", "cli"]
 triggers:
   - "remember this"
   - "recall what we decided"
@@ -15,9 +16,6 @@ triggers:
   - "sync Revolve"
   - "evidence loop"
   - "multi-agent memory"
-  - "fan-out memory"
-  - "fan-in memory"
-  - "coordinated memory"
 ---
 
 # Tree Ring Memory
@@ -33,6 +31,51 @@ Tree Ring Memory preserves meaningful agent learning like tree rings:
 - speculative future work stays as seeds
 - sensitive data is blocked, redacted, or kept out by default
 
+## Runtime Preflight
+
+Before running a Tree Ring command:
+
+1. Read project-local `.tree-ring/SKILL.md` and `.tree-ring/CLI.md` when they
+   exist. They describe the configured root and exact installed commands.
+2. Confirm that the local runtime is available:
+
+   ```bash
+   tree-ring --version
+   ```
+
+3. This package targets Tree Ring Memory CLI 0.14.0 or newer. If the command is
+   missing or older, do not invent results, edit shell configuration, or install
+   or upgrade software without the user's explicit permission. Explain the
+   limitation and point to the canonical install guide:
+   <https://github.com/TerminallyLazy/Tree-Ring-Memory#install>
+
+If the current host cannot execute a local shell or access project files, use
+this skill only as memory-lifecycle guidance. Do not claim that recall, capture,
+audit, activation, or forgetting occurred unless the corresponding command ran
+and its result was observed.
+
+## Agent Operating Loop
+
+Use this sequence for meaningful project work:
+
+1. Resolve the project root and read its local Tree Ring contract when present.
+2. Run the runtime preflight, then recall narrowly scoped, source-linked memory
+   before making a material decision or repeating a failure-prone workflow.
+3. Treat recalled memory as context, not authority. Recheck facts that may have
+   changed and defer to current source files, tests, policies, and user input.
+4. Do the work. Do not write memory merely because a session is active.
+5. At a natural checkpoint or closeout, capture only durable decisions,
+   corrections, validated lessons, warnings, or future seeds. Never store raw
+   transcripts, secrets, or sensitive data.
+6. Observe the command result and report the actual outcome. A proposed memory,
+   dry run, bridge file, or generated marker is not proof that a durable write,
+   sync, activation, correction, or deletion occurred.
+
+In a Coordinated store, do not attempt persistent writes without the required
+coordinator capability. If the capability is unavailable, provide a concise
+candidate memory for an authorized coordinator instead of claiming it was
+stored.
+
 ## When To Recall
 
 Recall memory before:
@@ -99,6 +142,98 @@ as a pointer only. Read the project-local `.tree-ring/SKILL.md` and
 Do not assume a global Tree Ring setup applies to the current repo unless the
 user explicitly configured it.
 
+## DOX Contract Flow
+
+When a project uses DOX-style `AGENTS.md` contracts:
+
+1. Read the applicable contract chain from the project root down to the working
+   directory before editing files. More specific child contracts may refine the
+   parent contract.
+2. Treat those current source files as authoritative. A recalled DOX summary is
+   only a navigation and continuity aid; it never overrides the live contract.
+3. Preview the adapter output first with
+   `tree-ring dox sync --source-root <path> --dry-run` and inspect every summary
+   and source reference.
+4. Persist only concise, useful summaries. In a Coordinated store, persistence
+   requires coordinator authority; dry-run discovery does not.
+5. Never use the adapter to rewrite a root or child `AGENTS.md`, copy whole
+   contract trees into memory, or weaken child instructions. Re-run the dry run
+   after a source contract changes and re-read the chain before the next edit.
+
+## Harness Activation
+
+For a new project, begin with the safe, project-local default:
+
+```bash
+tree-ring init
+tree-ring integrations status
+```
+
+Do not ask the user to copy a bridge or run `integrations link` for ordinary
+setup. `init` configures only safe project-local adapter material by creating
+absent final bridge and manifest paths. It never replaces or removes an existing
+entry, including during deactivation; contested entries stay untouched and
+report `needs-user-review`. A bridge, marker, generated skill, or successful
+`init` is not activation proof: `active` requires a fresh, matching receipt from
+a new session's scoped recall and safe context injection.
+Treat `configured-awaiting-proof`, `active-isolated`, `needs-trust`,
+`needs-project-mount`, `needs-plugin`, `needs-user-review`, `unsupported`,
+and `failed` as their exact non-active outcomes. Never say Hermes or another
+unverified runtime is active.
+
+If publication durability becomes indeterminate, do not delete or rewrite the
+published path. Preserve disk material, keep changed harnesses marked
+`needs-user-review` in the returned in-memory manifest, and leave any activation
+manifest already published on disk intact for explicit reconciliation.
+
+Pi trust is the user's decision: report `needs-trust` rather than changing
+global trust. Agent Zero is separate: `tree-ring init` writes only Tree Ring's
+passive Agent Zero binding with `needs-plugin`. The user installs/enables the
+compatible `tree_ring_memory` plugin and selects the mounted project; the plugin
+then owns its absolute, non-project `activation-capability.json` descriptor and
+passes it internally. Only descriptor-scoped plugin status can derive
+`configured-awaiting-proof`, and only its new-session preflight receipt can
+make the runtime `active`.
+
+Never create a generic marker, copy or hand-author that descriptor, set its
+internal transport, modify Agent Zero core, or call a different plugin store
+shared. A missing, invalid, disabled, or release-incompatible descriptor stays
+`needs-plugin`; a different reachable store is `active-isolated`; an
+unavailable root is `needs-project-mount`. A passive binding, source checkout,
+or stale bundled CLI is not installed capability.
+
+Receipts prove a privacy-safe preflight check, not durable memory creation or a
+security boundary. They exclude raw prompts, recalled content, secrets,
+sensitive values, paths, and coordinator capabilities. Shared-store claims are
+limited to same-host local-filesystem processes whose receipts match the
+canonical project `store_id`; they do not apply across hosts or network
+filesystems. For diagnostics use `tree-ring integrations status --verbose`;
+for advanced controlled work use `integrations activate --harness <id>
+--dry-run`, `integrations certify`, or `integrations deactivate --harness
+<id>`.
+
+## Certification Boundary
+
+For an installed Tree Ring runtime, use the self-contained CLI evidence paths:
+
+```bash
+tree-ring integrations certify --source-root .
+tree-ring recall-quality --source-root .
+```
+
+Harness certification is non-mutating and writes JSON/Markdown evidence under
+`target/tree-ring-certification/`; it does not activate a harness or prove that
+an agent used recalled context. Keep receipt-backed status as a separate gate.
+
+`sh scripts/certify-tree-ring.sh` is the full framework release suite. Run it
+only from a canonical Tree Ring Memory source checkout where that file, the Rust
+workspace, `install.sh`, fixtures, and build tooling are all present. Do not
+copy it into another project, download it automatically, or claim full release
+certification from the smaller installed-CLI checks. In the TUI, `/evidence
+refresh` only displays this external source-checkout command; it does not run
+certification. If the script is absent, report that boundary and use the
+self-contained CLI commands above when they fit the user's request.
+
 Evidence outcome mapping:
 
 - `promoted`: durable heartwood from supported evidence

TERMS.md

@@ -0,0 +1,13 @@
+# Tree Ring Memory Skill Terms of Use
+
+Effective August 23, 2026
+
+This Tree Ring Memory skill is open-source software distributed under the MIT
+License. It provides agent instructions only and is not a hosted memory,
+backup, access-control, medical-record, or legal-record service.
+
+The user is responsible for choosing what to store, maintaining backups,
+controlling local filesystem access, reviewing agent actions, and complying
+with applicable policies and law. The skill is provided without warranty to
+the maximum extent permitted by law. The limitations in the included MIT
+License apply.