@forwardimpact/outpost 3.0.3

package/templates/.claude/skills/extract-entities/references/templates-goals-priorities.md

@@ -0,0 +1,63 @@
+# Goal and Priority Templates
+
+Goals and Priorities are **never auto-created** by `extract-entities`. They are
+set deliberately by the user. These templates are for manual creation only —
+`extract-entities` and `hyprnote-process` only **link to** and **update progress
+on** existing notes.
+
+## Priorities
+
+```markdown
+# {Priority Name}
+
+## About
+{2-3 sentences: what this strategic direction means and why it matters}
+
+**Status:** {active|paused|retired}
+**Owner:** [[People/{Person}]]
+**Set:** {YYYY-MM-DD}
+
+## What this means
+{Bullet list of concrete implications — what does pursuing this priority look like?}
+
+## Goals
+{Time-bound targets that ladder to this priority — backlinks to knowledge/Goals/}
+
+## Projects
+- [[Projects/{Project}]] — {relationship}
+
+## Key facts
+{substantive facts only — leave empty if none}
+```
+
+## Goals
+
+```markdown
+# {Goal Name}
+
+## Info
+**Priority:** [[Priorities/{Priority}]]
+**Status:** {on track|at risk|off track|achieved|abandoned}
+**Owner:** [[People/{Person}]]
+**Target date:** {YYYY-MM-DD}
+**Set:** {YYYY-MM-DD}
+
+## Outcome
+{1-2 sentences: what measurable success looks like}
+
+## Blockers
+{Active Conditions that impede this goal — link to knowledge/Conditions/}
+- [[Conditions/{Condition}]] — {impact on this goal}
+
+## Projects
+- [[Projects/{Project}]] — {how it contributes}
+
+## Progress
+- **{YYYY-MM-DD}**: {Update on progress toward the outcome}
+
+## Key facts
+{substantive facts only — leave empty if none}
+
+## Risks
+{known risks to achieving this goal — leave empty if none}
+```

package/templates/.claude/skills/extract-entities/references/templates-people-orgs.md

@@ -0,0 +1,67 @@
+# People and Organization Templates
+
+## People
+
+```markdown
+# {Full Name}
+
+## Info
+**Role:** {role or inferred role with qualifier}
+**Organization:** [[Organizations/{organization}]]
+**Reports to:** [[People/{{Person}}]]
+**Email:** {email}
+**Aliases:** {comma-separated variants}
+**First met:** {YYYY-MM-DD}
+**Last seen:** {YYYY-MM-DD}
+
+## Summary
+{2-3 sentences: who they are, why you know them, what you're working on}
+
+## Connected to
+- [[Organizations/{Org}]] — works at
+- [[People/{Person}]] — {relationship}
+- [[Projects/{Project}]] — {role}
+
+## Activity
+- **{YYYY-MM-DD}** ({meeting|email|voice memo}): {Summary with [[Folder/Name]] links}
+
+## Key facts
+{substantive facts only — leave empty if none}
+
+## Open items
+{commitments and next steps only — leave empty if none}
+```
+
+## Organizations
+
+```markdown
+# {Organization Name}
+
+## Info
+**Type:** {company|team|institution}
+**Industry:** {industry}
+**Relationship:** {customer|prospect|partner|vendor}
+**Domain:** {primary email domain}
+**Aliases:** {comma-separated}
+**First met:** {YYYY-MM-DD}
+**Last seen:** {YYYY-MM-DD}
+
+## Summary
+{2-3 sentences}
+
+## People
+- [[People/{Person}]] — {role}
+
+## Contacts
+{for transactional contacts who don't get their own notes}
+
+## Projects
+- [[Projects/{Project}]] — {relationship}
+
+## Activity
+- **{YYYY-MM-DD}** ({type}): {Summary}
+
+## Key facts
+
+## Open items
+```

package/templates/.claude/skills/extract-entities/references/templates-projects-topics.md

@@ -0,0 +1,68 @@
+# Project and Topic Templates
+
+## Projects
+
+```markdown
+# {Project Name}
+
+## Info
+**Type:** {deal|product|initiative|hiring}
+**Status:** {active|planning|on hold|completed|cancelled}
+**Started:** {YYYY-MM-DD}
+**Last activity:** {YYYY-MM-DD}
+
+## Summary
+{2-3 sentences}
+
+## Goals
+- [[Goals/{Goal}]] — {how this project contributes}
+
+## People
+- [[People/{Person}]] — {role}
+
+## Organizations
+- [[Organizations/{Org}]] — {relationship}
+
+## Related
+- [[Priorities/{Priority}]] — {relationship}
+- [[Topics/{Topic}]] — {relationship}
+
+## Timeline
+**{YYYY-MM-DD}** ({type})
+{What happened with [[links]]}
+
+## Decisions
+- **{YYYY-MM-DD}**: {Decision}. {Rationale}.
+
+## Open items
+
+## Key facts
+```
+
+## Topics
+
+```markdown
+# {Topic Name}
+
+## About
+{1-2 sentences}
+
+**Keywords:** {comma-separated}
+**Aliases:** {other references}
+**First mentioned:** {YYYY-MM-DD}
+**Last mentioned:** {YYYY-MM-DD}
+
+## Related
+- [[People/{Person}]] — {relationship}
+- [[Organizations/{Org}]] — {relationship}
+
+## Log
+**{YYYY-MM-DD}** ({type}: {title})
+{Summary with [[links]]}
+
+## Decisions
+
+## Open items
+
+## Key facts
+```

package/templates/.claude/skills/extract-entities/scripts/state.mjs

@@ -0,0 +1,131 @@
+#!/usr/bin/env bun
+/**
+ * Manage graph_processed state for entity extraction.
+ *
+ * Tracks which source files (synced emails and calendar events) have already
+ * been processed by the extract-entities skill. The `check` command lists files
+ * that are new or changed since their last recorded hash; `update` marks files
+ * as processed by storing their current SHA-256 hash.
+ *
+ * State is persisted as a TSV file at
+ * ~/.cache/fit/outpost/state/graph_processed (path<TAB>hash).
+ */
+
+if (process.argv.includes("-h") || process.argv.includes("--help")) {
+  console.log(`state — manage graph_processed state for entity extraction
+
+Usage:
+  node scripts/state.mjs check                    List new/changed files
+  node scripts/state.mjs update <path> [<path>…]  Mark files as processed
+  node scripts/state.mjs -h|--help                 Show this help message
+
+State file: ~/.cache/fit/outpost/state/graph_processed (TSV: path<TAB>hash)`);
+  process.exit(0);
+}
+
+import { createHash } from "node:crypto";
+import {
+  existsSync,
+  mkdirSync,
+  readFileSync,
+  readdirSync,
+  statSync,
+  writeFileSync,
+} from "node:fs";
+import { homedir } from "node:os";
+import { join } from "node:path";
+
+const HOME = homedir();
+const STATE_FILE = join(HOME, ".cache/fit/outpost/state/graph_processed");
+const SOURCE_DIRS = [
+  join(HOME, ".cache/fit/outpost/apple_mail"),
+  join(HOME, ".cache/fit/outpost/apple_calendar"),
+  join(HOME, ".cache/fit/outpost/teams_chat"),
+];
+
+/** Compute SHA-256 hash of a file. */
+function fileHash(filePath) {
+  const data = readFileSync(filePath);
+  return createHash("sha256").update(data).digest("hex");
+}
+
+/** Load the state file into a Map of {path → hash}. */
+function loadState() {
+  const state = new Map();
+  if (!existsSync(STATE_FILE)) return state;
+  const text = readFileSync(STATE_FILE, "utf-8");
+  for (const line of text.split("\n")) {
+    if (!line) continue;
+    const idx = line.indexOf("\t");
+    if (idx === -1) continue;
+    state.set(line.slice(0, idx), line.slice(idx + 1));
+  }
+  return state;
+}
+
+/** Write the full state Map back to the state file. */
+function saveState(state) {
+  const dir = join(HOME, ".cache/fit/outpost/state");
+  mkdirSync(dir, { recursive: true });
+  const entries = [...state.entries()].sort((a, b) => a[0].localeCompare(b[0]));
+  const text = entries.length
+    ? entries.map(([p, h]) => `${p}\t${h}`).join("\n") + "\n"
+    : "";
+  writeFileSync(STATE_FILE, text);
+}
+
+/** Find source files that are new or have changed since last processing. */
+function check() {
+  const state = loadState();
+  const newFiles = [];
+  for (const dir of SOURCE_DIRS) {
+    if (!existsSync(dir)) continue;
+    for (const name of readdirSync(dir)) {
+      const filePath = join(dir, name);
+      const stat = statSync(filePath, { throwIfNoEntry: false });
+      if (!stat || !stat.isFile()) continue;
+      const h = fileHash(filePath);
+      if (state.get(filePath) !== h) {
+        newFiles.push(filePath);
+      }
+    }
+  }
+  newFiles.sort();
+  for (const f of newFiles) {
+    console.log(f);
+  }
+  return newFiles.length;
+}
+
+/** Mark files as processed by updating their hashes in state. */
+function update(filePaths) {
+  const state = loadState();
+  for (const fp of filePaths) {
+    if (!existsSync(fp)) {
+      console.error(`Warning: File not found: ${fp}`);
+      continue;
+    }
+    state.set(fp, fileHash(fp));
+  }
+  saveState(state);
+  console.log(`Updated ${filePaths.length} file(s) in graph state`);
+}
+
+// --- CLI ---
+
+const args = process.argv.slice(2);
+const cmd = args[0];
+
+if (cmd === "check") {
+  const count = check();
+  console.error(`\n${count} file(s) to process`);
+} else if (cmd === "update" && args.length >= 2) {
+  update(args.slice(1));
+} else {
+  console.error(
+    "Usage:\n" +
+      "  node scripts/state.mjs check\n" +
+      "  node scripts/state.mjs update <file-path> [<file-path> …]",
+  );
+  process.exit(1);
+}

package/templates/.claude/skills/hyprnote-follow/SKILL.md

@@ -0,0 +1,174 @@
+---
+name: hyprnote-follow
+description: >
+  Follow a live Hyprnote session in real-time, coaching the user through a
+  meeting or interview. Understands context from the session title, knowledge
+  base, and candidate pipeline. Provides talking points, flags gaps in
+  coverage, and suggests follow-up questions as the conversation unfolds.
+  Use when the user asks to follow, shadow, or coach them through a live meeting.
+---
+
+# Hyprnote Follow
+
+Follow a live Hyprnote recording, read the transcript as it grows, and coach the
+user through the meeting in real time. Gather knowledge-base context once before
+the session, then poll the transcript and provide actionable nudges as new
+content appears.
+
+## Trigger
+
+- The user asks to follow, shadow, or coach them through a live meeting.
+- "Follow my meeting", "coach me", "shadow this call".
+- The user starts a Hyprnote recording and wants real-time support.
+
+## Prerequisites
+
+- Hyprnote installed; sessions at
+  `~/Library/Application Support/hyprnote/sessions/`.
+- An active or about-to-start session.
+- Knowledge base populated (attendee / candidate context).
+
+## Inputs
+
+- Live `transcript.json` (growing during the session).
+- `knowledge/People/`, `knowledge/Candidates/{Name}/{brief,screening,panel}.md`,
+  `knowledge/Roles/`, `knowledge/Organizations/`, `knowledge/Projects/`.
+- `~/.cache/fit/outpost/apple_calendar/*.json` for context.
+
+## Outputs
+
+- Real-time coaching messages printed to the user.
+- **No files are written.** This skill is purely advisory.
+
+<do_confirm_checklist goal="Verify the follow session was useful and read-only">
+
+- [ ] Active session detected and confirmed with the user.
+- [ ] Meeting type classified from the title.
+- [ ] Knowledge-base context gathered for all attendees.
+- [ ] For interviews: standard expectations and screening focus areas loaded.
+- [ ] Coaching nudges were actionable and concise (1–3 lines each).
+- [ ] Coverage gaps tracked and surfaced before the meeting ended.
+- [ ] End-of-meeting detected; debrief provided.
+- [ ] Next steps offered (`req-assess` / `hyprnote-process`); user decided
+      whether to run them.
+- [ ] No files were modified during the session.
+
+</do_confirm_checklist>
+
+## Procedure
+
+### Phase 1 — Detect and connect
+
+#### 1. Find the active session
+
+```bash
+node .claude/skills/hyprnote-follow/scripts/follow.mjs --detect
+```
+
+Returns the most recently modified session, whether it's live, and its title. If
+nothing has been modified in the last 5 minutes, warn the user and ask whether
+to follow a specific session. If multiple sessions could be active, confirm with
+the user.
+
+#### 2. Read session metadata
+
+```bash
+node .claude/skills/hyprnote-follow/scripts/follow.mjs <session-id> --meta
+```
+
+Capture **title**, **created_at**, **participants**.
+
+#### 3. Classify the meeting type
+
+Title pattern → type → coaching focus:
+[references/meeting-types.md](references/meeting-types.md). The type drives
+Phase 2 context loading and Phase 3 dimensions.
+
+### Phase 2 — Gather context (once)
+
+#### 4. Resolve attendees
+
+Extract names from title and participant list. For each:
+
+```bash
+rg -l "{name}" knowledge/People/
+rg -l "{name}" knowledge/Candidates/
+```
+
+Read matching notes for role, organization, history, open items, prior
+interactions.
+
+#### 5. Load type-specific context
+
+**Interviews:** read `knowledge/Candidates/{Name}/{brief,screening,panel}.md`,
+look up the `Req` field's matching `knowledge/Roles/*.md`, and load standard
+expectations:
+
+```bash
+bunx fit-pathway job {discipline} {level} --track={track}
+```
+
+**General meetings:** read attendee People notes plus referenced
+Project/Organization notes. Check open tasks: `rg "{name}" knowledge/Tasks/`.
+
+#### 6. Build the coaching brief
+
+Synthesize gathered context into the pre-meeting brief format in
+[references/coaching.md](references/coaching.md#pre-meeting-brief-format) and
+print it to the user.
+
+### Phase 3 — Follow loop
+
+#### 7. Read new transcript content
+
+First read (no `--after`):
+
+```bash
+node .claude/skills/hyprnote-follow/scripts/follow.mjs <session-id>
+```
+
+Subsequent reads (pass the last word ID):
+
+```bash
+node .claude/skills/hyprnote-follow/scripts/follow.mjs <session-id> --after <last-word-id>
+```
+
+Returns JSON with grouped text segments, channel labels, and the next last-word
+ID.
+
+#### 8. Analyze new content
+
+Track the dimensions in
+[references/coaching.md](references/coaching.md#dimensions-to-track) — interview
+vs. general — and apply the [principles](references/coaching.md#principles) and
+[constraints](references/coaching.md#constraints).
+
+#### 9. Provide coaching
+
+Output only when actionable. Use the
+[coaching output formats](references/coaching.md#coaching-output-formats). Keep
+each nudge 1–3 lines.
+
+#### 10. Detect meeting end
+
+Watch for: farewells ("bye", "take care", "have a good day"), wrap-up phrases
+("that's all", "let's wrap", "I'll let you go"), or final thank-yous with no
+substantive follow-up. When detected, move to Phase 4.
+
+#### 11. Loop cadence
+
+**Do not use `sleep` or timed loops.** After each coaching output, read the next
+batch immediately — natural cadence emerges from read/analyze/output. On two
+consecutive empty reads, ask the user whether the meeting has ended.
+
+### Phase 4 — Wrap-up
+
+#### 12–13. Debrief
+
+Use the appropriate template in [references/debrief.md](references/debrief.md) —
+interview vs. general.
+
+#### 14. Offer next steps
+
+Offer (don't ask) the follow-ups in
+[references/debrief.md](references/debrief.md#next-step-offers-step-14).

package/templates/.claude/skills/hyprnote-follow/references/coaching.md

@@ -0,0 +1,118 @@
+# Coaching Reference
+
+Reference for `hyprnote-follow` Phases 2 and 3.
+
+## Pre-meeting brief format
+
+### For interviews
+
+```
+Following: {Title}
+Type: {Interview type}
+Candidate: {Name} — {current role} at {employer}
+Role: {target role} ({level}, {track})
+Screening: {recommendation from screening.md}
+
+Key areas to probe:
+- {Focus area 1 from screening/panel brief}
+- {Focus area 2}
+- {Focus area 3}
+
+Strengths to confirm:
+- {Strength 1}
+- {Strength 2}
+
+Watching for: {specific signals at this interview stage}
+```
+
+### For general meetings
+
+```
+Following: {Title}
+Attendees: {names with roles}
+
+Open items with {name}:
+- {item 1}
+- {item 2}
+
+Suggested topics:
+- {based on recent activity/projects}
+```
+
+## Dimensions to track
+
+### Interviews
+
+| Dimension               | What to watch for                                                                  |
+| ----------------------- | ---------------------------------------------------------------------------------- |
+| **Skills demonstrated** | Candidate shows (not just claims) standard skills                                  |
+| **Behaviour signals**   | Own the Outcome, Think in Systems, Communicate with Precision, Polymath, Curiosity |
+| **Level signals**       | Autonomy, scope, complexity handling, mentoring                                    |
+| **Coverage gaps**       | Focus areas from screening that haven't been touched yet                           |
+| **Red flags**           | Contradictions with CV, vague answers, deflection                                  |
+| **Time awareness**      | Are you running out of time with uncovered areas?                                  |
+
+### General meetings
+
+| Dimension             | What to watch for                                    |
+| --------------------- | ---------------------------------------------------- |
+| **Open items**        | Were pending items raised and addressed?             |
+| **Decisions**         | What was decided? Who owns next steps?               |
+| **New commitments**   | Track who committed to what                          |
+| **Unasked questions** | Topics the user should raise before the meeting ends |
+| **Tone/dynamics**     | Shifts in energy, resistance, enthusiasm             |
+
+## Coaching output formats
+
+Output **only when actionable**. Each nudge: 1–3 lines max.
+
+```
+Probe deeper: {Name} mentioned {topic} — ask for a specific example
+Gap: screening flagged {skill} as uncertain. Try: "{question}"
+Confirmed: {Name} demonstrated {skill} at {level} — "{brief quote}"
+Follow up: {Name} committed to {action} — pin this down with a deadline
+Time check: ~{N} minutes in. Still uncovered: {topic 1}, {topic 2}
+Capture: decision made on {topic} — {who} will {what}
+```
+
+## Principles
+
+### Interviews
+
+- **Standard-grounded.** Every nudge ties to standard skills/behaviours. Don't
+  coach on vibes.
+- **Evidence over claims.** Flag _claims_ vs. _demonstrations_; push for
+  specific examples.
+- **Coverage-aware.** Track which screening focus areas have been explored and
+  nudge before time runs out.
+- **Level-calibrated.** Know what the target level expects; flag
+  below-/above-level answers.
+- **Don't lead.** Suggest open-ended probes, never loaded or leading questions.
+
+### General meetings
+
+- **Agenda-aware.** Track coverage of known topics or open items.
+- **Decision-focused.** Flag decisions and capture owners.
+- **Commitment-tracking.** Note who said they'd do what; surface unresolved
+  commitments from prior meetings.
+- **Time-conscious.** Nudge when the meeting is running long and key topics
+  aren't raised.
+
+### Universal
+
+- **Be quiet when things are going well.** Speak only when actionable.
+- **Channel awareness.** Channel 0 = user; channel 1+ = guest(s). Reference what
+  the _guest_ said — the user already knows what they said.
+- **Never interrupt flow.** Hold nudges for the next natural pause.
+- **Respect the user.** Coach at the margin, don't patronise.
+
+## Constraints
+
+- **Read-only.** Never modify Hyprnote files, knowledge notes, or state.
+- **No post-processing during follow.** Don't run `extract-entities`,
+  `req-assess`, or `hyprnote-process`. Offer them after.
+- **Channel 0 = user.** Always.
+- **Transcription noise.** Live transcripts are messy. Don't coach from a single
+  ambiguous word — wait for context.
+- **Privacy.** Coaching may appear on a shared screen — keep nudges generic
+  enough that they don't reveal confidential assessment detail.

package/templates/.claude/skills/hyprnote-follow/references/debrief.md

@@ -0,0 +1,59 @@
+# Debrief Templates
+
+Reference for `hyprnote-follow` Phase 4 (Steps 12–13).
+
+## Interview debrief
+
+```markdown
+## Session Debrief: {Title}
+**Duration:** ~{N} minutes
+
+### Coverage
+| Focus Area | Status | Evidence |
+|---|---|---|
+| {area from screening} | Covered / Partial / Missed | {brief note} |
+
+### Signals Observed
+**Strengths:**
+- {skill/behaviour}: {what was demonstrated}
+
+**Concerns:**
+- {skill/behaviour}: {what was lacking or contradictory}
+
+**Level read:** {level assessment based on what was observed}
+
+### Suggested Next Steps
+- {e.g. "Schedule decomposition interview to probe {skill}"}
+- {e.g. "Flag {concern} for panel brief"}
+```
+
+## General-meeting debrief
+
+```markdown
+## Session Debrief: {Title}
+**Duration:** ~{N} minutes
+
+### Decisions Made
+- {decision 1}
+- {decision 2}
+
+### Action Items
+| Owner | Action | Deadline |
+|---|---|---|
+| {name} | {what} | {when, if mentioned} |
+
+### Open Items (still unresolved)
+- {item that wasn't addressed}
+
+### Key Takeaways
+- {1–2 sentence summary of what matters most}
+```
+
+## Next-step offers (Step 14)
+
+After the debrief, **offer** (don't ask) logical follow-ups:
+
+- **Interviews:** "I can run `req-assess` on this transcript when Hyprnote
+  finishes processing."
+- **Meetings:** "I can extract entities from this session into the knowledge
+  graph when ready."

package/templates/.claude/skills/hyprnote-follow/references/meeting-types.md

@@ -0,0 +1,14 @@
+# Meeting-Type Classification
+
+Reference for `hyprnote-follow` Phase 1, Step 3. Classify from the session
+title.
+
+| Title pattern                               | Type                        | Coaching focus                                                 |
+| ------------------------------------------- | --------------------------- | -------------------------------------------------------------- |
+| "Interview with {Name}", "Screening {Name}" | **Screening interview**     | Agent-aligned standard skills, behaviour probes, level signals |
+| "Decomposition interview with {Name}"       | **Decomposition interview** | Problem-solving approach, system thinking, trade-offs          |
+| "Technical assessment of {Name}"            | **Technical interview**     | Hands-on skill, architecture reasoning, code quality           |
+| "Panel with {Name}", "Second interview"     | **Panel interview**         | Business acumen, cultural fit, leadership signals              |
+| "Meeting with {Name}", "{Topic} meeting"    | **General meeting**         | Agenda coverage, action items, decision capture                |
+| "Recruitment call", "Recruitment strategy"  | **Recruitment planning**    | Pipeline status, role requirements, next steps                 |
+| Other                                       | **General**                 | Active listening, follow-up questions, key takeaways           |