Update to Matt Skills Curated

{
  "description": "Operate the agent's inner cognitive workspace for multi-step reasoning, long-horizon planning, deep debugging, and calibrated self-monitoring. Use when tackling complex multi-step problems, architecture design, long mechanical tasks requiring goal preservation, suspicious input audits, or when asked to think deeper or harder — even if they don't explicitly say \"J-space\". Do NOT use for single-step trivial lookups, basic formatting, or conversational small talk.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 121
    }
  ],
  "name": "j-space",
  "skill_md_contents": "---\nname: j-space\ndescription: \"Operate the agent's inner cognitive workspace for multi-step reasoning, long-horizon planning, deep debugging, and calibrated self-monitoring. Use when tackling complex multi-step problems, architecture design, long mechanical tasks requiring goal preservation, suspicious input audits, or when asked to think deeper or harder — even if they don't explicitly say \\\"J-space\\\". Do NOT use for single-step trivial lookups, basic formatting, or conversational small talk.\"\n---\n\n# J-Space: Inner Cognitive Workspace\n\nOperate the model's inner cognitive workspace (J-space) for multi-step chained reasoning, long-horizon planning, complex debugging, calibrated confidence, global consistency, and resistant self-monitoring.\n\n## The J-Space Premise\n\nYou do not only produce words; you also reason before — and without — immediately emitting them. J-space holds what you are **poised to say**: concepts you can hold in mind deliberately, reason with silently, and broadcast to downstream computations.\n\n---\n\n## Core Invariants\n\n1. **Restate Requirement Before Solving**: Before answering any non-trivial prompt, restate the requirement in one line in your own words to anchor the reasoning chain.\n2. **Three Explicit Registers**:\n   - **Inner**: Dense, private, chained reasoning track.\n   - **Ledger**: Durable state tracking (Goal, Settled, Open, Next). Checked at every seam.\n   - **Outer**: Clean, complete, professional communication delivered to the user.\n3. **The Gating Floor**: Classify the task before loading heavy machinery:\n   - `fast`: Single-step direct answers verifiable in one glance.\n   - `full`: 2–4 steps, single deliverable, verifiable in one reading.\n   - `loop`: Multi-stage, multi-file, multi-turn stateful work.\n4. **Seam Audits**: At every seam (sub-task completion, file write, tool call, deliverable checkpoint), audit the ledger before proceeding.\n5. **No Blind Completion**: Never mark a task finished without re-reading the initial goal line-by-line and verifying all acceptance criteria.\n\n---\n\n## Architecture & Map of Content (MOC)\n\n```\n[ Incoming Task ] ──► [ Restate Requirement ] ──► [ Task Gate: Fast / Full / Loop ]\n                             │\n                             ▼\n ┌──────────────────────────────────────────────────────────┐\n │ J-Space Cognitive Loop                                  │\n │   ├─ Inner Register (Chained intermediate reasoning)     │\n │   ├─ Ledger Register (Goal / Open / Settled / Next)      │\n │   └─ Seam Audit (Check invariants at each transition)    │\n └───────────────────────────┬──────────────────────────────┘\n                             │\n                             ▼\n                 [ Outer Register Delivery ]\n```\n\n| Functional Property | Cognitive Purpose | Failure Mode Prevented |\n|---|---|---|\n| **Capacity Selectivity** | Only 1–2 active concepts on stage at once | Attention thinning & overloaded context |\n| **Directed Focus** | Goal remains active through tedious middle | Goal evaporation during mechanical tasks |\n| **Deep Reasoning** | Intermediate bridges form before conclusions | Rationalizing unverified gut guesses |\n| **Self-Monitoring** | Calibrated confidence & error detection | Overconfident halluncinations |\n| **Seam Auditing** | State refreshed at every transition | Context drift over long execution turns |\n\n---\n\n## Step-by-Step Procedure (TWI)\n\n### Step 1: Awakening & Requirement Re-Encoding\n- **Action**: Restate the requirement in one line, in your own words, before writing code or making tool calls.\n- **Key Point**: Re-encoding buys back recurrence and anchors the goal representation.\n- **Why**: Skipping re-encoding leads to shallow interpretation of multi-constraint prompts.\n\n### Step 2: Task Gate Classification\n- **Action**: Classify into `fast`, `full`, or `loop` pass.\n- **Key Point**: If you cannot verify the answer in a single glance, it is never `fast`. Escalating costs nothing.\n- **Why**: Under-classifying complex tasks causes premature shortcutting.\n\n### Step 3: Operate the Three Registers\n- **Action**: Think in the inner register, record state in the ledger, and speak in the outer register.\n- **Key Point**: Keep the ledger updated at every seam: Goal, Settled items, Open questions, Next single action.\n- **Inline Checklist**:\n  - [ ] Requirement re-encoded\n  - [ ] Correct pass classified\n  - [ ] Ledger reflects current state\n  - [ ] Verification criteria stated explicitly\n\n### Step 4: Seam Refresh & Invariant Verification\n- **Action**: At every seam (file written, subagent returned, tool executed), check that invariants hold.\n- **Key Point**: If an approach fails, declare a marker, perform the bound corrective action, and settle before continuing.\n- **Why**: Continuing down a broken path wastes turns and produces compounding errors.\n\n### Step 5: Clean Outer Delivery\n- **Action**: Translate settled conclusions into clean outer-register deliverables.\n- **Key Point**: No private symbols or unstructured fragments in user-facing deliverables.\n- **Why**: Deliverables must be immediately usable, clear, and actionable.\n\n---\n\n## Anti-Rationalization Guardrails\n\n| Tempting Rationalization | Binding Rule | Engineering Rationale |\n|---|---|---|\n| *\"This is simple enough, skip the ledger.\"* | **Enforce the floor: multi-step work requires ledger tracking.** | Unsettled state drops out of context during long turns. |\n| *\"The first guess sounds right, skip the intermediate step.\"* | **Derive intermediate bridges before adopting conclusions.** | Pretrained fluency often produces plausible but flawed shortcuts. |\n| *\"I will clean up the verification at the very end.\"* | **Verify at every seam, not just at the end.** | Errors caught early require small fixes; late discoveries require total rewrites. |\n| *\"Confidence is always high across all steps.\"* | **Calibrate confidence per step.** | Uniform confidence indicates disabled self-monitoring. |\n"
}