skillwiki 0.10.22 → 0.10.23

package/dist/{index-projection-Q6WL76SO.js → index-projection-FRWTZB5Y.js}

@@ -4,7 +4,8 @@ import {
   UNMANAGED_START,
   renderRootIndex,
   writeRootIndexProjection
-} from "./chunk-54JVBNWK.js";
+} from "./chunk-QBZYEEBD.js";
+import "./chunk-6ZDKTNLA.js";
 import "./chunk-C2DKFJFA.js";
 export {
   UNMANAGED_END,

package/dist/skillwiki-mcp.js

@@ -1,10 +1,12 @@
 #!/usr/bin/env node
 import {
   runSkillwikiMcpStdio
-} from "./chunk-7YDWC6D3.js";
+} from "./chunk-SYKSL3JQ.js";
 import "./chunk-7I2TPIV5.js";
-import "./chunk-54JVBNWK.js";
+import "./chunk-QBZYEEBD.js";
 import "./chunk-PQG26AGJ.js";
+import "./chunk-QNTBNNEL.js";
+import "./chunk-6ZDKTNLA.js";
 import "./chunk-C2DKFJFA.js";
 
 // src/mcp-entry.ts

package/dist/sources-L2SQV63E.js

@@ -0,0 +1,10 @@
+#!/usr/bin/env node
+import {
+  runSourcesPending
+} from "./chunk-EQPU2BPM.js";
+import "./chunk-QNTBNNEL.js";
+import "./chunk-6ZDKTNLA.js";
+import "./chunk-C2DKFJFA.js";
+export {
+  runSourcesPending
+};

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "skillwiki",
-  "version": "0.10.22",
+  "version": "0.10.23",
   "type": "module",
   "bin": {
     "skillwiki": "dist/cli.js",

package/skills/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "skillwiki",
-  "version": "0.10.22",
+  "version": "0.10.23",
   "skills": "./",
   "description": "Project-aware Karpathy-style knowledge base for Claude Code: 19 prompt-only skills (wiki-*, proj-*, using-skillwiki) backed by the deterministic `skillwiki` CLI.",
   "author": {

package/skills/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "skillwiki",
-  "version": "0.10.22",
+  "version": "0.10.23",
   "description": "Project-aware Karpathy-style knowledge base for Codex with 19 prompt-only skills backed by the deterministic skillwiki CLI.",
   "author": {
     "name": "karlorz",

package/skills/README.md

@@ -3,7 +3,7 @@
 Prompt-only Markdown skills for Claude Code. Installed via `skillwiki install`
 or the Claude/Codex/Antigravity plugin packaging paths.
 
-Current package inventory: **18 skills**.
+Current package inventory: **19 skills**.
 
 Publication policy: new or updated typed-knowledge and meta pages must use
 `skillwiki page publish` from a temporary draft, inspect its dry-run, and add
@@ -11,6 +11,16 @@ Publication policy: new or updated typed-knowledge and meta pages must use
 typed path or separately edit its index/log entries; immutable raw sources and
 non-typed project work retain their existing workflows.
 
+Raw lifecycle policy: completed raw captures are byte-immutable. Attended,
+state-bound structural commands may preserve-move sources under
+`raw/archived/` or `raw/duplicates/`; ordinary remove refuses raw, and permanent
+disposal requires `skillwiki sources dispose` for one explicit exact target.
+Web Clipper captures enter asynchronously through the init-installed
+`_Templates/web-clipper/llm-wiki-clippings.json`; inspect them with
+`skillwiki sources pending` or the separate `query --include-pending` channel.
+`raw/assets/**` is a flexible stable pool with explicit vault-root embeds, not a
+fixed papers/transcripts taxonomy.
+
 | Namespace | Skills |
 |---|---|
 | `wiki-*` | `wiki-init`, `wiki-ingest`, `wiki-query`, `wiki-lint`, `wiki-crystallize`, `wiki-audit`, `wiki-archive`, `wiki-reingest`, `wiki-adapter-prd`, `wiki-add-task`, `wiki-sync`, `wiki-canvas`, `wiki-gate-plan-mode` |
@@ -48,8 +58,9 @@ Before a managed vault mutation, invoke the managed SkillWiki command while the
 
 - typed pages: `skillwiki page publish <draft> <vault> --target <path>` then the same command with `--write`
 - archive: `skillwiki archive <path> <vault>`
+- pending inventory/query: `skillwiki sources pending <vault>` and `skillwiki query "..." <vault> --include-pending`
+- raw disposition/disposal: `skillwiki sources disposition ...` / attended `skillwiki sources dispose ...`
 - ad-hoc structural log: `skillwiki log-append <vault> --content '<entry>'` (Release A dual-write) or event materialization (Release B)
 - project/root index: `skillwiki project-index <slug> <vault> --apply` and `skillwiki index rebuild <vault> --write` only through managed commands
 - log projection: `skillwiki log materialize <vault> [--write]`
 - paired projections: `skillwiki projections materialize <vault> [--write]`
-

package/skills/agents/wiki-add-task.md

@@ -39,7 +39,7 @@ You are a quick-capture agent specializing in writing ad-hoc captures to `raw/tr
    project: "[[{slug}]]"  # omit if no project
    ---
    ```
-   No `sha256` — ad-hoc captures are mutable working notes.
+   `sha256` may be omitted, but the completed capture is still immutable evidence.
 6. **Write body:** `# {type}: {text}` then the text content.
 7. **Cross-reference (optional).** If project slug provided, verify `projects/{slug}/` exists. Append one-line reference to project compound notes.
 8. **Log.** Append to `{vault}/log.md`: `## [YYYY-MM-DD] capture | [type]: [text (first 60 chars)]`.
@@ -60,6 +60,7 @@ Return:
 **Forbidden:**
 - Creating an `inbox/` directory
 - Appending to existing capture files
+- Editing or correcting an existing raw transcript; create a new capture or maintained work-item note
 - Creating a full work item (that's proj-work's job)
 - Writing to Layer 2 or Layer 3 locations (captures are Layer 1)
 - Writing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets to the vault

package/skills/agents/wiki-archive.md

@@ -1,34 +1,33 @@
 ---
 name: wiki-archive
-description: Use this agent when archiving superseded typed-knowledge pages during automated maintenance cycles. Typical triggers include dev-loop IDLE DISCOVERY cleanup, retiring pages replaced by newer versions, or post-reingest old-raw archival. See "When to invoke" in the agent body for worked scenarios.
+description: Use this agent for typed-page archival and report-only raw lifecycle previews; raw apply requires attended exact-target approval.
 model: sonnet
 color: yellow
 tools: ["Read", "Edit", "Bash", "Grep", "Glob"]
 ---
 
-You are a vault archivist specializing in safely retiring typed-knowledge pages. You move pages to `_archive/`, remove index entries, and verify no broken links remain. You operate autonomously during maintenance cycles — archive targets are specified in your task prompt.
+You are a vault archivist specializing in safely retiring typed pages and previewing raw preserve-moves. Automated maintenance may apply typed archival, but raw structural apply is never autonomous.
 
 ## When to invoke
 
 - **Page superseded.** A new version of a concept/entity page exists and the old one should be retired.
-- **N9 reingest archival.** Raw files are being re-ingested due to content drift — old raw must be archived.
+- **Raw lifecycle preview.** An exact raw source may need preserve-archive after attended review.
 - **Cleanup cycle.** Dev-loop spawns you to archive pages flagged during lint/audit.
 
 **Your Core Responsibilities:**
 1. Run `skillwiki archive <page>` to move the page to `_archive/`
 2. Verify no ghost entries remain with `skillwiki index-check`
 3. Check for broken wikilinks from other pages referencing the archived page
-4. For raw file archiving (N9 protocol): update all `^[raw/...]` citations that reference the old path
+4. For raw targets, produce the dry-run only and surface the state-bound approval token to an attended parent/operator
 
 **Execution Process:**
 
 1. **Resolve vault.** Run `skillwiki path`. If NO_VAULT_CONFIGURED, report failure and STOP.
 2. **Identify target.** The page to archive is specified in your task prompt.
-3. **Run archive.** Execute `skillwiki archive <page> <vault>`. Read the JSON output. If non-zero, report and STOP.
+3. **Run archive.** For typed pages, execute `skillwiki archive <page> <vault>`. For an exact raw path, execute preview only; do not add `--apply`. Raw destination must be `raw/archived/<category>/...`, with exact bytes preserved and no asset movement.
 4. **Verify index.** Run `skillwiki index-check <vault>`. Confirm no ghost entries remain.
 5. **Check broken links.** Run `skillwiki lint <vault>`. If other pages still wikilink to the archived page, update them to point to the replacement or remove the stale link.
-6. **N9 raw archiving.** When archiving a `raw/` file: update ALL `^[raw/...]` citation markers and `sources:` frontmatter in referencing pages. Change `raw/articles/foo.md` to `_archive/raw/articles/foo.md`. Verify with `skillwiki audit`.
-7. **Log.** Append to `{vault}/log.md`: `## [{date}] archive | {relPath} → _archive/{subdir}/`.
+6. **Raw handoff.** Report exact source/destination, complete-file hash, citation impact, and approval token. Only an attended explicitly invoked workflow may rerun with `--apply --approve <token>`.
 
 **Output Format:**
 Return:
@@ -43,7 +42,7 @@ Return:
 - Page not found or already archived
 
 **Forbidden:**
-- Archiving `raw/` files outside N9 Reingest Protocol
-- Archiving without updating citation markers for raw files
+- Applying any raw archive from an automated/background agent
+- Rewriting raw bytes/frontmatter, moving raw outside `raw/`, writing `_archive/raw/`, or moving referenced assets
 - Deleting files (archive moves, never deletes)
 - Preserving live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets by archive-only handling

package/skills/agents/wiki-audit.md

@@ -25,7 +25,7 @@ You are a vault provenance auditor specializing in verifying that every `^[raw/.
 1. **Resolve vault.** Run `skillwiki path`. If NO_VAULT_CONFIGURED, report failure and STOP.
 2. **Run audit.** Execute `skillwiki audit <page>` for each target page. If no page specified, audit all typed-knowledge pages (entities/, concepts/, comparisons/, queries/, meta/). Read the JSON report.
 3. **Reason over findings:**
-   - **Unresolved markers:** The `^[raw/...]` path does not resolve to an existing file. Suggest ingesting the missing source or correcting the citation path.
+   - **Unresolved markers:** Resolve against active `raw/`, `raw/archived/`, `raw/duplicates/`, and relocation history; legacy `_archive/raw/` is read-only compatibility. Distinguish genuinely missing evidence from a stale historical address.
    - **Unused sources:** A source is listed in `sources:` frontmatter but never cited in the body. Suggest adding a body citation or removing from `sources:`.
    - **Missing from sources:** A body citation lacks a corresponding `sources:` entry. Suggest adding to `sources:`.
 4. **Append summary.** Write one entry to `{vault}/log.md` summarizing audit findings and suggested follow-ups.
@@ -44,4 +44,5 @@ Return a structured summary:
 **Forbidden:**
 - Auto-applying suggested fixes (audit is observation-only — do not edit pages)
 - Modifying `sources:` frontmatter or body citations
+- Rewriting raw evidence
 - Printing or writing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets in audit summaries

package/skills/agents/wiki-lint.md

@@ -27,7 +27,7 @@ You are a vault health inspector specializing in running `skillwiki health` and
 2. **Run health or lint summary.** For whole-system health, execute `skillwiki health <vault> --out /tmp/skillwiki-health.json --no-fail`. For lint-only maintenance, execute `skillwiki lint <vault> --summary`. Read the JSON envelope. Treat `skillwiki doctor` as setup/runtime diagnostics only.
 3. **Drill into details only when needed.** If capped examples are insufficient, run the bucket's `details_command` or `skillwiki lint <vault> --only <bucket>`.
 4. **Reason over findings.** Group by severity. For each kind of finding, suggest concrete next actions. If the CLI was recently updated, new checks may flag pre-existing pages — treat these as legitimate findings, not false positives.
-5. **Sensitive content.** Treat `sensitive_content` as a security error. Drill down with `skillwiki lint <vault> --only sensitive_content --human`. Redaction is allowed as a security exception to raw immutability only through `skillwiki lint <vault> --fix --only sensitive_content`; never print the secret value in the report.
+5. **Sensitive content.** Treat `sensitive_content` as a security error. Drill down with `skillwiki lint <vault> --only sensitive_content --human`; never print the secret value. Raw findings remain report/quarantine-only: do not redact or rewrite existing raw content/frontmatter.
 6. **Log rotation.** If `log_rotate_needed` is present, note that user consent is required — do NOT auto-rotate.
 7. **Post-migration check.** If content was recently migrated, note whether broken_wikilinks count decreased after re-running `skillwiki lint <vault> --summary`. Remaining broken links for migrated content indicate pages still referencing moved files.
 8. **Optional summary.** Append one entry to `{vault}/log.md` with the lint counts (errors/warnings/info) and a timestamp only when explicitly requested.
@@ -49,5 +49,6 @@ Return a structured summary:
 **Forbidden:**
 - Auto-rotating logs without user consent
 - Auto-updating sha256 fields
+- Rewriting or redacting raw evidence
 - Modifying any page beyond an explicitly requested lint summary entry in `log.md`
 - Printing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets in findings or summaries

package/skills/agents/wiki-query.md

@@ -25,6 +25,7 @@ You are a vault search and synthesis specialist using E2 4-signal ranking to fin
 
 1. **Resolve vault.** Run `skillwiki path`. If NO_VAULT_CONFIGURED, report failure and STOP.
 2. **Determine scope.** From task prompt: full vault, current project, or project+concepts.
+   - Default to typed knowledge. If the prompt explicitly asks for new/recent/fresh/clipped/raw/pending/undigested/unprocessed/not-yet-integrated material, run `skillwiki sources pending` and `skillwiki query "..." --include-pending`.
 3. **Refresh graph.** If `.skillwiki/graph.json` missing or >24h old: `skillwiki graph build <vault>`.
 4. **Compute overlap.** `skillwiki overlap <vault>`.
 5. **Score candidates.** Apply 4 signals:
@@ -33,7 +34,7 @@ You are a vault search and synthesis specialist using E2 4-signal ranking to fin
    - Adamic-Adar: 1.5× (from graph output)
    - Type affinity: 1.0×
 6. **Read top candidates.** Read frontmatter + body of highest-scored pages.
-7. **Synthesize answer.** Compose with explicit citations to candidate pages using `^[page-path]` markers.
+7. **Synthesize answer.** Compose with explicit citations to candidate pages using `^[page-path]` markers. Keep `pending_sources` separate from ranked typed results, label them pending evidence, and never let them alter typed scores/order. Weak typed results may prompt a suggestion, not silent widening.
 8. **Sensitive content guard.** Before filing a query or comparison page, scan the generated body for live credentials, access keys, tokens, passwords, cookies, bearer headers, or private keys. Redact before writing. If the answer depends on preserving a live secret, STOP and ask for a redacted source or explicit rotation/remediation direction.
 9. **Optional file.** If the task asks to persist, first run `skillwiki page publish --help`. If it is unavailable, fail closed and leave the result unpublished. Otherwise:
    1. Resolve the vault and create a temporary directory outside the vault.
@@ -75,4 +76,5 @@ Return:
 - Directly creating a final typed page or separately editing `index.md` or `log.md` for its publication
 - Skipping graph refresh when graph.json is missing
 - Accepting wiki claims without filesystem verification
+- Treating pending raw captures as integrated typed knowledge or silently widening into raw
 - Writing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets to the vault

package/skills/agents/wiki-reingest.md

@@ -6,7 +6,7 @@ color: yellow
 tools: ["Read", "Write", "Edit", "Bash", "Grep", "Glob"]
 ---
 
-You are a drift detection specialist running `skillwiki drift` and processing results. When sources have changed since ingestion, you archive old raw files and re-ingest updated content following N9 immutability protocol. You operate autonomously during maintenance cycles.
+You are a drift detection specialist running `skillwiki drift` and processing results. Automated maintenance is report-only for raw evidence; attended reingest creates a new capture and preserve-archives the old one.
 
 ## When to invoke
 
@@ -17,7 +17,7 @@ You are a drift detection specialist running `skillwiki drift` and processing re
 **Your Core Responsibilities:**
 1. Run `skillwiki drift` to detect changed sources
 2. Present findings grouped by status (drifted, fetch_failed, unchanged)
-3. For each drifted source: archive old raw, ingest new content, update citations
+3. For each drifted source: report the proposed new-capture + attended preserve-archive workflow
 4. Log the results
 
 **Execution Process:**
@@ -29,16 +29,16 @@ You are a drift detection specialist running `skillwiki drift` and processing re
    - **fetch_failed:** Could not re-fetch. Note error details.
    - **unchanged:** No action needed.
 4. **Process each drifted source:**
-   a. Archive old raw: `skillwiki archive <raw-path>`
+   a. In background/headless mode, stop after reporting. Never run raw structural apply.
    b. Re-fetch content. If it contains live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets, STOP and ask for a redacted source or explicit rotation/remediation direction.
-   c. Write as new raw file with updated sha256
-   d. Update all concept/entity pages citing the old source: change `^[raw/...]` markers and `sources:` to reference the new path
-   e. Verify with `skillwiki audit` that no broken markers remain
+   c. In an attended explicitly approved workflow, write the fetched content as a new raw file with updated sha256 and verify it.
+   d. Preview `skillwiki archive <exact-old-raw-path>`; apply only with the live `--apply --approve <token>` authorization. Old bytes remain under `raw/archived/`.
+   e. Update maintained citations to the new capture where editorially appropriate and verify with `skillwiki audit`.
 5. **Log.** Append to `{vault}/log.md`: scanned count, drifted count, re-ingested count, skipped count.
 
 **N9 Compliance:**
 Raw files are immutable. Never modify an existing raw file. Instead:
-- Archive old raw → `_archive/raw/`
+- Preserve-archive old raw → `raw/archived/<category>/`
 - Create new raw with updated content and new sha256
 - This preserves full provenance history
 
@@ -58,6 +58,7 @@ Return:
 
 **Forbidden:**
 - Modifying files in `raw/` directly (N9)
-- Re-ingesting without archiving old raw first
+- Applying raw reingest/archive from scheduled or headless maintenance
+- Writing new archives to `_archive/raw/`
 - Updating citations without running `skillwiki audit` to verify
 - Writing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets to the vault

package/skills/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@skillwiki/skills",
-  "version": "0.10.22",
+  "version": "0.10.23",
   "private": true,
   "files": [
     "wiki-*",

package/skills/skills/using-skillwiki/SKILL.md

@@ -37,14 +37,17 @@ Invoke a skillwiki skill when the user:
 
 ## Vault Structure
 A skillwiki vault has three layers. The canonical architecture lives in `SCHEMA.md` at the vault root — read it before creating any new directories.
-**Layer 1 — Raw (`raw/`):** Immutable source material. Never modify after ingest. `raw/transcripts/` doubles as the ad-hoc capture point for meeting notes and unprocessed ideas.
+**Layer 1 — Raw (`raw/`):** Immutable evidence. Existing content/frontmatter is never autonomously rewritten or removed. Attended structural workflows may rename, relocate, archive, or deduplicate only when exact bytes remain somewhere under `raw/`. Permanent disposal requires explicit exact-target user intent. `raw/transcripts/` is the ad-hoc capture point for meeting notes and unprocessed ideas.
 ```
 raw/
-├── articles/    # Web articles, clippings
-├── papers/      # PDFs, arxiv papers
-├── transcripts/ # Meeting notes, interviews, ad-hoc captures
-└── assets/      # Images, diagrams referenced by sources
+├── articles/             # Active Web articles and clippings
+├── papers/               # Active PDFs and papers
+├── transcripts/          # Active meeting notes and ad-hoc captures
+├── assets/               # Stable flexible asset pool; no fixed internal taxonomy
+├── archived/{articles,papers,transcripts}/
+└── duplicates/{articles,papers,transcripts}/
 ```
+Use explicit vault-root asset embeds such as `![[raw/assets/example/diagram.png]]`. Agents may choose flat or URL-friendly nested paths. Once an immutable capture references an asset, that path freezes; routine source archive/dedup never moves the asset. Remote images remain external dependencies unless separately captured.
 Raw frontmatter:
 ```yaml
 ---
@@ -94,6 +97,9 @@ Also mirror these pointers in vault-presync / vault-sync-status skills when oper
 
 - typed pages: `skillwiki page publish <draft> <vault> --target <path>` then the same command with `--write`
 - archive: `skillwiki archive <path> <vault>`
+- pending source inventory: `skillwiki sources pending <vault>`
+- editorial disposition: `skillwiki sources disposition <exact-raw-path> <vault> ...`
+- exceptional raw disposal: `skillwiki sources dispose <exact-raw-path> <vault> --reason ...` then attended `--write --approve <token>`
 - ad-hoc structural log: `skillwiki log-append <vault> --content '<entry>'` (Release A dual-write) or event materialization (Release B)
 - project/root index: `skillwiki project-index <slug> <vault> --apply` and `skillwiki index rebuild <vault> --write` only through managed commands
 - log projection: `skillwiki log materialize <vault> [--write]`
@@ -125,20 +131,20 @@ The vault is shared across hosts, so host-local absolute paths are not durable s
 | Entry | When | What happens |
 |-------|------|-------------|
 | `/wiki-add-task <text>` | You're in a Claude session | Creates `raw/transcripts/YYYY-MM-DD-{type}-{slug}.md` with ad-hoc capture frontmatter |
-| Filesystem drop | You're NOT in a Claude session (Obsidian, editor, sync) | Create/edit any `.md` file in `raw/transcripts/` — dev-loop discovers it on next cycle |
+| Filesystem drop | You're NOT in a Claude session (Obsidian, editor, sync) | Create a new `.md` file in `raw/transcripts/` — dev-loop discovers it on next cycle; do not edit it after capture |
 | Dev-loop discovery | Automatic, next cycle | Scans `raw/transcripts/` for new files since last cycle, surfaces as claimable work |
 
 ## Skill Map
 | Skill | When to Invoke |
 |-------|----------------|
-| `wiki-init` | Bootstrap a new vault — SCHEMA.md, index.md, log.md, ~/.skillwiki/.env |
+| `wiki-init` | Bootstrap a vault and install `_Templates/web-clipper/llm-wiki-clippings.json` plus import guidance |
 | `wiki-ingest` | Convert URLs, files, or pasted text into typed-knowledge pages |
-| `wiki-query` | Search the vault and synthesize an answer with ranked results |
+| `wiki-query` | Search typed knowledge by default; explicitly requested fresh/raw evidence uses the separate pending channel |
 | `wiki-lint` | Vault health and lint checks; use `health` for whole-system reports and `lint --summary` for bounded lint buckets |
 | `wiki-crystallize` | Distill the current working session into a typed-knowledge page |
 | `wiki-audit` | Verify raw provenance references and source frontmatter integrity |
-| `wiki-archive` | Archive a typed-knowledge page — move to `_archive/`, remove from index (writes delete-intent) |
-| `wiki-remove` | Hard-delete a path with delete-intent tombstone; failsafe-git if CLI missing |
+| `wiki-archive` | Archive typed pages, or attended preserve-move exact raw sources under `raw/archived/` |
+| `wiki-remove` | Remove maintained pages; exact raw disposal uses the separate attended `sources dispose` flow |
 | `wiki-reingest` | Detect drift in raw sources (sha256 comparison) and re-ingest updated content |
 | `wiki-add-task` | Quick-capture ideas, bugs, tasks, notes into `raw/transcripts/` without leaving the current workflow |
 | `wiki-adapter-prd` | Map foreign PRD formats (CodeStable, RFC, AIDE, Hermes) into vault pages |
@@ -174,8 +180,8 @@ Use `prd_layer` + `prd_pipeline` from `.claude/dev-loop.config.md` as source of
 
 ## CLI Backbone
 All skills are backed by the `skillwiki` CLI — a deterministic tool with no LLM calls. It handles path resolution, config management, validation, health reporting, and linting. Skills invoke it via Bash for the mechanical parts and use Claude for the creative parts.
-Key CLI subcommands: `init`, `health`, `lint`, `config`, `doctor`, `path`, `lang`, `install`, `fleet context`, `fleet validate`, `graph build`, `archive`, `remove`, `drift`, `compound`, `tag-sync`, `tag reconcile`, `page publish`, `sync status`, `seed`, `stale`, `observe`, `canvas generate`.
-Run `skillwiki health <vault> --out /tmp/skillwiki-health.json --no-fail` for a bounded whole-system report that includes doctor, lint, vault-sync, query-readiness, source-freshness, risk flags, and self-check coverage. Run `skillwiki lint <vault> --summary` for lint-only bucket counts with capped examples and details commands. Run `skillwiki doctor` to diagnose setup/runtime issues only. Run `skillwiki config list` to see current configuration.
+Key CLI subcommands: `init`, `health`, `lint`, `config`, `doctor`, `path`, `lang`, `install`, `fleet context`, `fleet validate`, `graph build`, `query`, `sources pending`, `sources disposition`, `sources dispose`, `archive`, `remove`, `drift`, `dedup`, `compound`, `tag-sync`, `tag reconcile`, `page publish`, `sync status`, `seed`, `stale`, `observe`, `canvas generate`.
+Run `skillwiki health <vault> --out /tmp/skillwiki-health.json --no-fail` for a bounded whole-system report that includes the nonblocking source-lifecycle backlog. Pending captures are informational and do not make health fail. Run `skillwiki lint <vault> --summary` for lint-only bucket counts with capped examples and details commands. Run `skillwiki doctor` to diagnose setup/runtime issues only. Run `skillwiki config list` to see current configuration.
 
 ## Runtime Host Context and Fleet Freshness
 Resolve the active project vault with `skillwiki path` first. Then pass that exact path to `skillwiki --human fleet context <vault>` for host identity and safety guidance. `fleet context` is authoritative for host identity. It overrides stale injected SessionStart context, remembered workspace context, and prior conversation summaries. `fleet context` is local and network-free; it reports `identity_status`, resolver trace, warnings, and the fact that remote freshness was not checked.
@@ -214,7 +220,7 @@ Before asking questions or running scheduled maintenance, resolve the session ki
 ## Typical Workflow
 1. **Init** (`wiki-init`) — create vault, set domain and taxonomy
 2. **Ingest** (`wiki-ingest`) — add sources, build pages
-3. **Query** (`wiki-query`) — search and synthesize answers
+3. **Query** (`wiki-query`) — search typed knowledge; explicitly include pending captures for fresh/raw intent
 4. **Lint** (`wiki-lint`) — periodic health checks
 5. **Crystallize** (`wiki-crystallize`) — save session insights as pages
 6. **Audit** (`wiki-audit`) — verify source integrity

package/skills/skills/wiki-add-task/SKILL.md

@@ -8,7 +8,7 @@ Capture ad-hoc ideas, bugs, tasks, and notes into the vault. Three entry points
 |-------|------|-------------|
 | `/wiki-add-task <text>` | You're in a Claude Code session (NOT Hermes compact) | Creates `raw/transcripts/YYYY-MM-DD-{type}-{slug}.md` with ad-hoc capture frontmatter |
 | Filesystem drop | Hermes Agent compact mode (no slash commands available) | Same as above — create `.md` in `raw/transcripts/`, dev-loop discovers it |
-| Filesystem drop | You're NOT in a Claude session (Obsidian, editor, sync) | Create any `.md` file in `raw/transcripts/` using the vault template — dev-loop discovers it on next cycle |
+| Filesystem drop | You're NOT in a Claude session (Obsidian, editor, sync) | Create a new `.md` file in `raw/transcripts/` using the vault template — dev-loop discovers it on next cycle |
 | Dev-loop discovery | Automatic, next cycle | Scans `raw/transcripts/` for new files since last cycle, surfaces as claimable work |
 **Path Rule:** Captures ALWAYS go to `$(skillwiki path)/raw/transcripts/` (Layer 1). Never under `projects/{slug}/raw/` — that violates SCHEMA.md Layer 1 immutability.
 ### Exception: Explicit project task requests
@@ -47,7 +47,7 @@ project: "[[{slug}]]"
 - If a `project` slug was provided, set `project: "[[slug]]"`.
 - If no project, omit the `project` field entirely.
 - `source_url` is null (these are locally originated captures).
-- No `sha256` — ad-hoc captures are mutable working notes, not immutable sources.
+- `sha256` may be omitted for locally originated captures, but the completed capture is still immutable evidence. Corrections create a new capture or a maintained work-item note; never rewrite the existing transcript.
 5. **Write body.** Below the frontmatter, write:
 ```markdown
 # {type}: {text}
@@ -79,7 +79,7 @@ Fix the template mismatch between wiki-add-task and the vault template.
 ```
 The `kind` field uses the capture type and must be one of: `idea`, `bug`, `task`, `note` (plus the existing `postmortem`, `session-log`, `meeting-notes`, `other` for non-capture raw sources).
 The `project` and `kind` fields can be set independently — they do not require `work_item`. The `work_item` field is only used when the raw source is directly tied to a project work item (set by `proj-work`).
-Ad-hoc captures omit `sha256` — they are mutable working notes, not immutable sources. The `sha256` field is reserved for ingested raw sources that require integrity verification.
+Ad-hoc captures may omit `sha256`; omission does not grant mutation authority. Once created, the transcript's content and frontmatter are immutable. The `sha256` field remains required for ingest pipelines that provide integrity verification.
 ## Stop conditions
 - `skillwiki path` returns NO_VAULT_CONFIGURED.
 - No `text` provided (prompt user once, then stop).
@@ -88,6 +88,7 @@ Ad-hoc captures omit `sha256` — they are mutable working notes, not immutable
 ## Forbidden
 - Creating an `inbox/` directory. All captures go to `raw/transcripts/`.
 - Appending to existing capture files — each capture gets its own file.
+- Editing or correcting an existing raw transcript; create a new capture or maintained work-item note instead.
 - Creating a work item — this is capture-only. Use `proj-work` for full work items.
 - Writing to any Layer 2 or Layer 3 location. Captures are Layer 1 (raw).
 - Writing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets to the vault.

package/skills/skills/wiki-archive/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: wiki-archive
-description: Archive a superseded typed-knowledge page. Moves page to _archive/, removes from index.md, logs the action.
+description: Archive a superseded typed page or preserve-move an exact raw source through the attended lifecycle workflow.
 ---
 
 # wiki-archive
@@ -8,7 +8,8 @@ description: Archive a superseded typed-knowledge page. Moves page to _archive/,
 ## When This Skill Activates
 
 - User wants to retire, supersede, or remove a typed-knowledge page from active use.
-- A page has been replaced by a newer version and should be kept for reference but excluded from lint and queries.
+- A page has been replaced by a newer version and should be kept for reference but excluded from active use.
+- The user explicitly requests a structural archive of one exact raw source while preserving its bytes under `raw/`.
 
 ## Output language
 
@@ -20,29 +21,32 @@ Standard four reads (SCHEMA, index, log, project context if applicable).
 
 ## Probe
 
-Same matrix as `using-skillwiki` → **CLI probe and failsafe**: PRIMARY (`skillwiki archive`), FAILSAFE-GIT (hand tombstone + `git mv` to `_archive/` + push), or FAIL CLOSED.
+Same matrix as `using-skillwiki` → **CLI probe and failsafe**. FAILSAFE-GIT is permitted for typed pages only. Raw archive requires the SkillWiki attended approval flow; if it is unavailable, fail closed.
 
 ## Steps (PRIMARY)
 
 0. Resolve vault: `skillwiki path` and `skillwiki lang`.
-1. Identify the target page. Confirm with the user which page to archive (show full relPath).
-2. Run `skillwiki archive <page> [vault]`. On a vault-sync leaf host where S3 stale originals must be pruned, use `skillwiki archive <page> [vault] --remote seaweed-wiki:cloud/wiki --remote-delete --max-remote-deletes 1` only when that remote path deletion is explicitly intended. Read the JSON output. Successful archive **always** writes a live-path tombstone under `meta/delete-intents/` so snapshot cannot resurrect the live key.
-3. Verify with `skillwiki index-check [vault]` — confirm no ghost entries remain.
-4. Run `skillwiki lint [vault]` — check for broken wikilinks from other pages that still reference the archived page. If found, update those pages to point to the replacement or remove the stale link.
-5. **Raw file archiving (N9 Reingest Protocol only):** When archiving a `raw/` file due to content drift, update ALL `^[raw/...]` citation markers and `sources:` frontmatter entries that reference the old path. Change `raw/articles/foo.md` to `_archive/raw/articles/foo.md` in every referencing page. Verify with `skillwiki audit` that no broken markers remain.
-6. Append a `log.md` entry: `## [{date}] archive | {relPath} → _archive/{subdir}/`.
-7. Commit/push so tombstone + archive land on private `main`.
+1. Identify the target page and show its full vault-relative path. Raw targets must be exact `raw/...` paths; basename inference is refused.
+2. For a typed page, run `skillwiki archive <page> [vault]` using the normal managed mutation workflow.
+3. For a raw article, paper, or transcript:
+   1. Run `skillwiki archive <exact-raw-path> [vault]` and inspect the dry-run, destination, complete-file hash, citation impact, and state-bound approval token.
+   2. If the user approves that exact live plan, run the identical command with `--apply --approve <token>`.
+   3. The destination is `raw/archived/<category>/...`. The command copies exclusively, verifies complete-file SHA-256, retires the old address, records append-only relocation history, and leaves the raw bytes unchanged.
+   4. Maintained citations may be rewritten to the new address; historical addresses also resolve through relocation history. Legacy `_archive/raw/` is read-compatible only and is never a new-write destination.
+4. On a vault-sync leaf where remote stale-path pruning is explicitly intended, add `--remote ... --remote-delete --max-remote-deletes 1`. This prunes only the retired address; the preserved raw destination remains.
+5. Verify with `skillwiki index-check [vault]`, `skillwiki lint [vault]`, and `skillwiki audit <referencing-page>` where applicable.
+6. Commit/push through the normal vault-sync workflow so the tombstone, preserved destination, and relocation event land together.
 
 ## FAILSAFE-GIT (no skillwiki)
 
-1. Confirm path. Write `meta/delete-intents/<slug>.json` with `action: "archive"`, `source: "failsafe-git"`, schema `vault-delete-intent/v1`.
-2. `git mv` live path → `_archive/<same relPath>` (create dirs as needed); update `index.md` if typed.
-3. Commit with `Delete-Intent` / `Delete-Source: failsafe-git` trailers; push to private `main`.
-4. Optional single-path `rclone deletefile` for the **live** path if remote configured.
+1. Confirm the target is not under `raw/`. Raw preserve-moves fail closed without the CLI transaction.
+2. Write `meta/delete-intents/<slug>.json` with `action: "archive"`, `source: "failsafe-git"`, schema `vault-delete-intent/v1`.
+3. `git mv` the typed page to `_archive/<same relPath>`; update `index.md`.
+4. Commit with `Delete-Intent` / `Delete-Source: failsafe-git` trailers; push to private `main`.
 
 ## Reversibility
 
-Archiving is locally reversible: move the file back from `_archive/` to its original directory, re-add the wikilink entry to `index.md`, and **delete the matching tombstone** under `meta/delete-intents/`. If `--remote-delete` was used, the stale active-path object is pruned from the remote after the archive move, but the archived copy remains and a restore will republish the active path on the next push once the tombstone is cleared.
+Typed archiving is locally reversible from `_archive/`. Raw archive is reversible only through another attended preserve-move that keeps exact bytes under `raw/`; do not hand-edit or recreate the source at its old address.
 
 ## Stop conditions
 
@@ -52,8 +56,9 @@ Archiving is locally reversible: move the file back from `_archive/` to its orig
 
 ## Forbidden
 
-- Archiving `raw/` files outside the N9 Reingest Protocol (raw is immutable except during content-drift reingestion).
-- Archiving raw files without updating all `^[raw/...]` citation markers that reference them.
-- Archiving without user confirmation.
+- Rewriting raw content or frontmatter during archive.
+- Moving raw evidence outside `raw/`, including new writes to `_archive/raw/`.
+- Moving `raw/assets/**` as a side effect of source archive; referenced asset paths stay frozen.
+- Applying a raw archive without a fresh state-bound token and attended explicit invocation.
 - Deleting local vault files with bare `rm` / bare `git rm` without a delete-intent tombstone (causes snapshot resurrection).
 - Remote stale-path pruning only via explicit `skillwiki archive --remote ... --remote-delete` (PRIMARY) or bounded single-path rclone in FAILSAFE-GIT.

package/skills/skills/wiki-audit/SKILL.md

@@ -21,7 +21,8 @@ Standard four reads.
 0. **Resolve vault and language.** Run `skillwiki path` (fail if NO_VAULT_CONFIGURED) and `skillwiki lang`.
 1. `skillwiki audit <page>`. Read the JSON report.
 2. Reason over the report:
-   - For each unresolved marker: suggest ingesting the missing source or correcting the path.
+   - Resolve citations against active `raw/`, `raw/archived/`, `raw/duplicates/`, and append-only relocation history. Legacy `_archive/raw/` is read-compatible only, not a valid new destination.
+   - For each unresolved marker: distinguish a genuinely missing target from a stale historical path, then suggest ingesting the missing source or correcting maintained-page metadata.
    - For each `unused_sources` entry: suggest adding a body marker or removing from `sources:`.
    - For each `missing_from_sources` entry: suggest adding to `sources:`.
 3. Append one `log.md` entry summarizing the audit and any suggested follow-ups.
@@ -31,3 +32,4 @@ None — audit always completes.
 
 ## Forbidden
 - Auto-applying suggested fixes (audit is observation-only).
+- Rewriting raw evidence while repairing an audit finding.

package/skills/skills/wiki-init/SKILL.md

@@ -22,7 +22,18 @@ None for the first run.
 3. Propose a 10–15 tag taxonomy tailored to the domain. Confirm or accept the user's revision.
 4. Ask the language question: "What language should generated page prose use? Default is `en`. Aliases like `chinese-traditional` or `zh-Hant` are accepted."
 5. Run `skillwiki init --target <dir> --domain "<answer>" --taxonomy "<comma list>" --lang "<lang>"`.
-6. **Suggest first sources.** Propose 3–5 initial sources (URLs, papers, articles) appropriate to the domain. Prompt the user to provide the first one to ingest, then hand off to wiki-ingest.
+6. Report the installed Web Clipper assets at
+   `_Templates/web-clipper/llm-wiki-clippings.json` and
+   `_Templates/web-clipper/readme.txt`. Tell the user to open Obsidian Web
+   Clipper Settings, import the JSON, and repeat the import for every browser
+   profile that should capture into the vault. Re-init preserves existing
+   customized copies, including short files.
+7. Explain that the generic template preserves `{{content}}`; remote HTTP(S)
+   images remain external dependencies. An attended local-asset capture may
+   choose any URL-friendly path under `raw/assets/`, but it must write the asset,
+   emit an explicit vault-qualified `![[raw/assets/...]]` embed, verify
+   resolution/preview, and only then finalize the immutable raw note.
+8. **Suggest first sources.** Propose 3–5 initial sources (URLs, papers, articles) appropriate to the domain. Prompt the user to provide the first one to ingest, then hand off to wiki-ingest.
 
 ## Stop conditions
 

package/skills/skills/wiki-lint/SKILL.md

@@ -15,7 +15,7 @@ Standard four reads.
 2. For lint-only maintenance, run `skillwiki lint <vault> --summary`. This returns bounded bucket counts, capped examples, and `details_command` hints without full item arrays.
 3. Drill into important buckets with `skillwiki lint <vault> --only <bucket>` when examples are insufficient for remediation.
 4. Reason over findings; present grouped by severity with concrete suggested actions per kind. If the CLI was recently updated with new lint checks, re-running lint on the full vault may flag pre-existing pages that predate the new rule — treat these as legitimate findings, not false positives.
-5. Treat `sensitive_content` as a security error. Drill down with `skillwiki lint <vault> --only sensitive_content --human`. Redaction is allowed as a security exception to raw immutability only through `skillwiki lint <vault> --fix --only sensitive_content`; never print the secret value in the report.
+5. Treat `sensitive_content` as a security error. Drill down with `skillwiki lint <vault> --only sensitive_content --human`; never print the secret value. Raw findings are report/quarantine-only: `--fix` must not redact, rewrite, or recompute existing raw evidence. Maintained-page repairs may proceed only through their approved write workflow.
 6. If `log_rotate_needed` is present and the user consents, run `skillwiki log-rotate <vault> --apply`. Otherwise leave alone.
 7. **Post-migration verification**: If the user recently migrated content (e.g., moved entity/concept pages to another vault), re-run `skillwiki lint <vault> --summary` and verify that broken_wikilinks count decreased. Remaining broken links for migrated content indicate pages still referencing the moved files — these should be cleaned up (remove citations or migrate the referencing pages too).
 8. Append a `log.md` entry summarizing lint counts only when the user asked to record the maintenance result. Do not log routine `health` reports by default.
@@ -24,5 +24,6 @@ None — lint reports all findings even on per-page errors.
 ## Forbidden
 - Auto-rotating logs.
 - Auto-updating sha256 fields.
+- Rewriting or redacting existing raw content/frontmatter, even for a sensitive-content finding.
 - Modifying any page beyond a user-approved lint summary entry in `log.md`.
 - Printing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets in findings or summaries.

package/skills/skills/wiki-query/SKILL.md

@@ -13,6 +13,8 @@ Standard four reads (SCHEMA, index, log, project context if applicable).
 ## Steps
 0. **Resolve vault and language.** Run `skillwiki path` (fail if NO_VAULT_CONFIGURED) and `skillwiki lang`.
 1. **Determine scope.** Ask the user once if ambiguous: vault | current project | project+concepts.
+   - Ordinary questions are typed-knowledge-first. Run `skillwiki query "<text>" [vault]` without widening into raw captures.
+   - Explicit freshness language — `new`, `recent`, `fresh`, `clipped`, `raw`, `pending`, `undigested`, `unprocessed`, `not yet ingested`, or `not yet integrated` — requests the pending evidence channel. Run `skillwiki sources pending [vault]` for inventory and `skillwiki query "<text>" [vault] --include-pending` for synthesis support.
 2. **Refresh graph.** If `.skillwiki/graph.json` is missing or older than 24h: `skillwiki graph build <vault>`.
 3. **Compute overlap.** `skillwiki overlap <vault>`.
 4. **Score candidates** in prompt using the 4 signals:
@@ -22,6 +24,9 @@ Standard four reads (SCHEMA, index, log, project context if applicable).
 - Type affinity: 1.0×
 5. **Read top candidates** in full (frontmatter + body).
 6. **Synthesize answer** with explicit citations to the candidate pages.
+   - When pending evidence was requested, keep the CLI's `pending_sources` separate from ranked typed `results`. Pending captures never change typed scores or ordering.
+   - Label raw captures as **pending evidence**, not established vault knowledge. Explain that they have not yet been distilled into a maintained page.
+   - Weak typed results may justify suggesting a pending lookup, but never silently broaden the query.
 7. **Sensitive content guard.** Before filing any query or comparison page, scan the generated body for live credentials, access keys, tokens, passwords, cookies, bearer headers, or private keys. Redact before writing. If the answer depends on preserving a live secret, STOP and ask for a redacted source or explicit rotation/remediation direction.
 8. **Optional file.** If the user accepts, first run `skillwiki page publish --help`. If it is unavailable, fail closed and leave the result unpublished. Otherwise:
    1. Resolve the vault and create a temporary directory outside the vault.
@@ -52,4 +57,6 @@ When a wiki page (especially a work item `tasks.md`) claims that fixes were appl
 - Filing without the `skillwiki page publish` dry-run passing.
 - Directly creating a final typed page or separately editing `index.md` or `log.md` for its publication.
 - Skipping the orientation reads even for "quick" queries.
+- Treating a pending raw capture as integrated typed knowledge, or mixing it into typed ranking.
+- Silently searching pending sources when the user did not request fresh/raw/unprocessed material.
 - Writing live credentials, access keys, tokens, passwords, cookies, bearer headers, private keys, or other authenticating secrets to the vault.