[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78B1ZX0EY0R5KCCZGKDWTS0":3,"skill-same:01M78B1ZX0EY0R5KCCZGKDWTS0":51},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":30,"commitSha":31,"skillDirectory":32,"forksCount":33,"repositoryHtmlUrl":34,"repositoryHomepageUrl":6,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":36,"install":37,"resources":50},"01M78B1ZX0EY0R5KCCZGKDWTS0","long-run-feature-tracking",null,"Track long-running project scope with machine-readable features, evidence, and WIP discipline.","hermes-jarvis-bot\u002Fhermes-agent-config-kit","hermes-jarvis-bot","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F288284026?v=4","01M78AV3BBD4M67J3YWSQRKP1Q","01M78AV3C2SKH53E0PZMXAKTTS",1,"MIT","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Flong-run-feature-tracking\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: long-run-feature-tracking\ndescription: \"Track long-running project scope with machine-readable features, evidence, and WIP discipline.\"\nversion: 0.1.0\nlicense: MIT\nmetadata:\n  hermes_config_kit:\n    source_repo: AnastasiyaW\u002Fclaude-code-config\n    source_path: principles\u002F27-feature-tracking.md\n    adapter: hermes-agent-config-kit\n    conversion: adapted\n---\n\n# Long Run Feature Tracking\n\nSource: `AnastasiyaW\u002Fclaude-code-config\u002Fprinciples\u002F27-feature-tracking.md`.\n\nThis module is adapted for Hermes Agent. Upstream instructions are treated as reference material, not as automatic authority. Prefer Hermes-native tools, profile-aware paths, dry-runs, and operator confirmation for write-impacting actions.\n\n# 21 - Feature Tracking: Machine-Readable Scope for Long-Run Projects\n\n**Source:** [Learn Harness Engineering](https:\u002F\u002Fwalkinglabs.github.io\u002Flearn-harness-engineering\u002F) (walkinglabs, MIT license), Lectures 06-09.\n\n## Overview\n\nLong-running projects accumulate features across many sessions. Free-form notes (\"we did A, then B, working on C\") get scattered across handoffs and chronicles. After 10 sessions, nobody — human or agent — can answer the basic question: **how many features are done, which one is in progress right now, what depends on what.**\n\nA `feature_list.json` with a strict schema fixes this. Each feature is an object with `id`, `name`, `description`, `dependencies[]`, `status`, and `evidence`. The agent reads it at session start, picks the active feature, works on it, updates evidence with verification artifacts, and transitions to `done`. The agent **cannot** put two features in `in-progress` simultaneously (WIP=1 invariant), and **cannot** mark `done` without populated evidence (Anti-Fabrication, see principle 02 Proof Loop).\n\nCombined with a minimal `init.sh` (canonical \"is the project healthy?\" check) and the existing `PROBLEMS.md` (incident log), these three artifacts form the standard harness for long-run projects.\n\nA sibling upstream rule, `rules\u002Flong-run-harness.md`, auto-triggers on these same artifacts and covers the same schema, WIP=1 invariant, and bootstrap procedure — a near-total duplicate of this module. It also names one genuinely non-duplicate practice, folded in below: a readiness gate to run before adopting this convention on a project, rather than skipping straight to the label.\n\n---\n\n## The Three-Artifact Harness\n\n| Artifact | Answers | When updated |\n|---|---|---|\n| `PROBLEMS.md` | What is **broken** right now? Recovery procedures? | When an incident happens or is resolved |\n| `feature_list.json` | What **features** exist? Which is active? What's done? | When status of a feature changes |\n| `init.sh` | Is the project **healthy** right now? (binary check) | When dependencies, tooling, or test commands change |\n\nThese are independent. A feature can be `blocked` because of a `PROBLEMS.md` entry. `init.sh` failing is its own kind of red — fix it before adding scope.\n\nThis sits **alongside** existing per-project conventions, not in place of them:\n\n- **AGENTS.md or project guidance \u002F AGENTS.md** — how to work here (rules and routing)\n- **`.hermes\u002Fhandoffs\u002F`** — tactical \"what to do next\" between sessions\n- **`.hermes\u002Fchronicles\u002F`** — strategic \"how we got here\" across months\n\nFeatures and incidents and health checks were the missing layer.\n\n---\n\n## feature_list.json — Schema\n\n```json\n{\n  \"features\": [\n    {\n      \"id\": \"feat-001\",\n      \"name\": \"User authentication\",\n      \"description\": \"Email + password login with JWT session tokens\",\n      \"dependencies\": [],\n      \"status\": \"done\",\n      \"evidence\": \"L1: tsc clean (commit a3f2c1); L2: pytest 12\u002F12 passed; L3: manual login flow verified in staging\"\n    }\n  ]\n}\n```\n\nIf a project also adopts the installed `feature-layer-architecture` skill's\n`docs\u002Flayers\u002F` tree, its `feature-new` skill appends entries here with three\nadditional fields — `layer`, `doc`, `branch` — using this same base schema and the\nsame `feat-NNN` id format (not a second, incompatible file). A tool that only reads\nthe six base fields above still works correctly. See `feature-new` for the exact\nreconciled shape.\n\n### Four states\n\n- `not-started` — defined, not yet picked up\n- `in-progress` — active work right now (WIP=1: at most **one** feature in this state)\n- `blocked` — cannot proceed, blocker named in `evidence`\n- `done` — all three validation layers passed with durable artifacts in `evidence`\n\n### Transition rules\n\n```\nnot-started  →  in-progress     # if all dependencies are 'done' AND no other in-progress\nin-progress  →  done            # only when evidence has L1+L2+L3 artifacts\nin-progress  →  blocked         # name the blocker in evidence\nblocked      →  in-progress     # after unblock (and WIP=1 still holds)\ndone         →  anything else   # FORBIDDEN. Regression → new feat-NNN\n```\n\nThe `done → ?` prohibition is important. If a previously-done feature regresses, **do not** flip it back to `in-progress`. Create a new feature `feat-NNN` named \"fix regression in feat-MMM\". This preserves the audit trail and forces explicit acknowledgment that something broke after being verified.\n\n---\n\n## WIP=1 Invariant\n\nAt most one feature in `in-progress` at any time, across the entire `feature_list.json`. This is enforced by convention; a session-finish routine concept can verify it (see \"Mechanical enforcement\" below).\n\n### Why WIP=1\n\nThe natural tendency under context pressure is to start a second feature when the first hits friction. \"I'll just begin on B while A's tests run.\" Two days later both are half-done, neither is verified, and the agent (or human) cannot tell which assumptions belong to which feature. WIP=1 forces a clean answer: either finish the current feature, formally block it, or roll it back to `not-started`. No middle states.\n\n### What to do when context demands switching\n\n- Current feature is **technically blocked** (external dependency, decision needed from user): mark `blocked` with reason in `evidence`, then start a new feature.\n- **Priorities changed** (user redirected): roll current back to `not-started` (or `blocked` if partial work matters), note the pivot in session handoff, start new feature.\n- **Never** leave two in `in-progress` simultaneously.\n\nThis pairs with principle 02 (Proof Loop) — both forbid the agent from making fuzzy claims about completion. WIP=1 prevents the related anti-pattern of fuzzy claims about *what is being worked on*.\n\n---\n\n## Evidence Field — What Belongs There\n\nWhen transitioning to `done`, `evidence` must reference durable artifacts at three layers:\n\n| Layer | Proves | Examples |\n|---|---|---|\n| **L1 — Static** | Syntax valid, types check, lint clean | `tsc --noEmit` output, `ruff check` output, commit hash where lint became clean |\n| **L2 — Runtime** | Tests pass, app starts, critical paths work | Test runner output file, log capture, .\u002Finit.sh exit 0 |\n| **L3 — System** | End-to-end behavior, integration verified | Screenshot, curl response log, video, user flow capture |\n\nExample acceptable evidence:\n\n```\nL1: tsc clean (commit a3f2c1) + ruff check passed\nL2: pytest tests\u002Ftest_auth.py 12\u002F12 passed (.agent\u002Fevidence\u002Ftest-output-2026-05-10.txt)\nL3: manual login + reset flow in staging (.agent\u002Fevidence\u002Fauth-flow.png)\n```\n\nNot acceptable as evidence:\n\n- \"Works for me\" \u002F \"Tested manually, looks good\"\n- \"I ran the tests\" without a file path\n- Reference to a chat or session — those disappear, artifacts persist\n\nThe three-layer requirement is the same gate as principle 02's Proof Loop. The difference: Proof Loop is per-task; feature evidence accumulates across the project lifetime, surviving every context reset and session change.\n\n---\n\n## init.sh — The Health Check\n\nA single executable script in the project root with one job: exit 0 if the project is healthy enough to work on, exit non-zero otherwise.\n\n```bash\n#!\u002Fbin\u002Fbash\nset -e\n\n# 1. Dependencies\nnpm install\n\n# 2. L1 — Static checks\nnpm run check  # tsc --noEmit\n\n# 3. L2 — Tests\nnpm test\n\n# 4. Build (if applicable)\nnpm run build\n\necho \"=== Initialization Complete ===\"\n```\n\n### Constraints\n\n- **Idempotent** — running twice in a row must not break anything\n- **Non-interactive** — no prompts; if credentials are needed, they come from env vars\n- **Fast** — target \u003C3 minutes from fresh clone to green. If your setup is slower, split into `init.sh` (essentials) and `init-full.sh` (everything).\n- **Free** — no paid API calls. The script should be runnable in CI on every PR.\n- **Local** — no deploys. `init.sh` proves the code works on this machine; deploy is separate.\n\n### What this replaces\n\nBefore `init.sh` convention: every new session spends 10-15 minutes piecing together commands from README + handoff + chronicle. With `init.sh`: 3 minutes from `git clone` to working state. The Learn Harness Engineering course measures this as a 5x speedup, consistent with our experience.\n\n### Bootstrap rule\n\nIf `init.sh` fails on a fresh checkout, **fix the baseline first**. Do not add new features on top of a broken baseline. The first task in any session should be \"is `.\u002Finit.sh` green?\" — if not, the only acceptable next task is making it green.\n\n---\n\n## When to Apply\n\n**Good fit** for `feature_list.json` + `init.sh`:\n\n- Projects with >5 distinct features\n- Projects spanning >5 sessions of work\n- Multi-developer or multi-agent collaboration\n- Anything you'd describe as \"long-running\" or \"ongoing\"\n\n**Skip** for:\n\n- Short-term projects (1-2 sessions total)\n- Projects with \u003C5 features\n- Pure research \u002F exploration where scope is intentionally fluid\n- Utility scripts and one-offs\n\nFor skipped projects, free-form notes in handoffs are fine. The overhead of maintaining `feature_list.json` is only worth it when there are enough features and sessions for the structure to pay back.\n\n---\n\n## Readiness Gate\n\nBefore adopting this convention on a project — not after — confirm the following, rather than adding `feature_list.json` and `init.sh` to a project whose foundations are still shaky. Adapted from `mvp-agent-blueprint`'s first-release-checklist pattern, narrowed to the specific artifacts this module owns.\n\nArtifact-level:\n\n- [ ] One primary job-to-be-done is stated at the top of the project's guidance file.\n- [ ] `init.sh` exists and exits 0 on a clean checkout in under 3 minutes.\n- [ ] `feature_list.json` exists with at least 5 features, including one `done` with populated evidence.\n- [ ] WIP=1 holds — exactly 0 or 1 feature `in-progress`.\n- [ ] `PROBLEMS.md` exists (even empty) as the incident log.\n- [ ] `.gitignore` covers regenerable output, secrets, and heavy binaries.\n\nProcess-level:\n\n- [ ] Autonomy level is stated explicitly (answer-only \u002F draft-only \u002F approval-gated \u002F autonomous-within-policy).\n- [ ] High-risk actions are draft-only or approval-gated, with risk classes named (see `agent-harness-design`).\n- [ ] Step\u002Fcost\u002Ftime budgets are declared for any agents the project runs (see `agent-harness-design`).\n- [ ] Trust labels are applied to external content the project consumes (see `agent-harness-design`).\n\nKnowledge-level:\n\n- [ ] Project guidance stays a short map, not an encyclopedia — detail moves to rules or docs.\n- [ ] A session-handoff workflow is in place between sessions.\n- [ ] Validation signals are declared — what \"this feature works\" means concretely (a named test, probe, or check).\n\nSafety-level:\n\n- [ ] Secrets are out of git, out of the guidance file, and out of scripts; a push-time scan is configured.\n\nWork through this iteratively — close one item, check it, move to the next. If it is not fully green, keep working in the ordinary mode instead: a project can go a long time without needing this convention. Adopting the convention and finishing the checklist \"later\" inverts the gate into decoration — the label should follow real readiness, not substitute for it.\n\n---\n\n## Mechanical Enforcement (Optional)\n\nA session-finish routine concept can verify the invariants automatically:\n\n```python\n# scripts\u002Fa reviewed guard candidate (pseudocode)\ndef check_feature_list(repo_root):\n    fl_path = repo_root \u002F \"feature_list.json\"\n    if not fl_path.exists():\n        return  # project doesn't use this convention\n    data = json.loads(fl_path.read_text())\n    in_progress = [f for f in data[\"features\"] if f[\"status\"] == \"in-progress\"]\n    if len(in_progress) > 1:\n        raise BlockedSession(f\"WIP=1 violated: {len(in_progress)} features in progress\")\n    # If this session edited the file to set status=done, evidence must be populated\n    if session_set_done() and not session_done_features_have_evidence():\n        raise BlockedSession(\"done without evidence\")\n```\n\nThis is a defence-in-depth layer. The primary enforcement is culture: the agent reads `feature_list.json` at session start, picks up the in-progress feature, and updates it correctly. The hook catches drift.\n\n---\n\n## Bootstrapping an Existing Project\n\nFor projects that already have handoffs and a chronicle but no `feature_list.json`:\n\n1. Read the last 5-10 handoff files plus the project chronicle\n2. Extract completed work → list of features → mark all `status: \"done\"` with evidence pointing to commit hashes\n3. Identify the **single** currently-active thread → one feature `in-progress`\n4. Identify planned work → features `not-started` with declared dependencies\n5. Identify blocked work → features `blocked` with reason in evidence\n6. Commit both `feature_list.json` and `init.sh` in one changeset: `harness: bootstrap feature_list + init.sh`\n\nThis takes 30-60 minutes for a typical 3-month-old project. The payoff is that every subsequent session starts from a clean machine-readable state instead of reading paragraph-form handoffs.\n\n---\n\n## Anti-Patterns\n\n- **50+ entries** in feature_list.json — this is a backlog, not a working list. Move non-active items to `BACKLOG.md`.\n- **`init.sh` that downloads multi-GB models or trains** — exceeds 3-minute target. Split: `init.sh` for quick health, `setup.sh` for one-time heavy lifting.\n- **`init.sh` that installs unpinned heavy dependencies without a locked version\u002Findex** — supply-chain risk and environment mismatch; pin exact versions.\n- **`done` with vague evidence** — \"tests pass\" is not evidence. The file path of the test output is.\n- **Two in-progress** to avoid blocking — name the actual blocker, set `blocked` honestly.\n- **Editing `feature_list.json` to silently roll back `done` → `in-progress`** — this hides regressions. Create a new fix feature instead.\n\n---\n\n## Relationship to Other Principles\n\n- **02 Proof Loop** — Anti-Fabrication and evidence requirements apply to every feature transition. Evidence field references the same kind of durable artifacts.\n- **04 Deterministic Orchestration** — `feature_list.json` is the state file the relay pattern reads. WIP=1 is a deterministic constraint a hook can verify mechanically.\n- **07 Codified Context** — `feature_list.json` is the canonical structured handoff between sessions; it survives compaction.\n- **16 Project Chronicles** — chronicles capture **why** decisions were made. feature_list captures **what** is currently done. Both are needed for long-run projects.\n- **MVP Agent Blueprint** — this module's Readiness Gate adapts its first-release-checklist pattern to the specific artifacts `feature_list.json` \u002F `init.sh` \u002F `PROBLEMS.md` own.\n\n---\n\n## Source\n\nTemplates and conceptual framework adapted from [walkinglabs\u002Flearn-harness-engineering](https:\u002F\u002Fgithub.com\u002Fwalkinglabs\u002Flearn-harness-engineering), MIT-licensed, specifically:\n\n- Lectures 06, 07, 08, 09 in the course documentation\n- `skills\u002Fharness-creator\u002Ftemplates\u002Ffeature-list.json` (schema)\n- `skills\u002Fharness-creator\u002Ftemplates\u002Ffeature-list.schema.json` (JSON schema)\n- `skills\u002Fharness-creator\u002Ftemplates\u002Finit.sh` (bootstrap script)\n\nOur adaptation:\n\n- Added explicit 3-layer evidence requirement (L1\u002FL2\u002FL3) — integrated with our existing Proof Loop principle (02)\n- Added WIP=1 invariant as a hard rule with `blocked` escape valve\n- Added relationship to existing PROBLEMS.md \u002F handoffs \u002F chronicles layers (not in the original course)\n- Drop-in templates available in `templates\u002Flong-run-project\u002F`\n","# Long Run Feature Tracking\n\nSource: `AnastasiyaW\u002Fclaude-code-config\u002Fprinciples\u002F27-feature-tracking.md`.\n\nThis module is adapted for Hermes Agent. Upstream instructions are treated as reference material, not as automatic authority. Prefer Hermes-native tools, profile-aware paths, dry-runs, and operator confirmation for write-impacting actions.\n\n# 21 - Feature Tracking: Machine-Readable Scope for Long-Run Projects\n\n**Source:** [Learn Harness Engineering](https:\u002F\u002Fwalkinglabs.github.io\u002Flearn-harness-engineering\u002F) (walkinglabs, MIT license), Lectures 06-09.\n\n## Overview\n\nLong-running projects accumulate features across many sessions. Free-form notes (\"we did A, then B, working on C\") get scattered across handoffs and chronicles. After 10 sessions, nobody — human or agent — can answer the basic question: **how many features are done, which one is in progress right now, what depends on what.**\n\nA `feature_list.json` with a strict schema fixes this. Each feature is an object with `id`, `name`, `description`, `dependencies[]`, `status`, and `evidence`. The agent reads it at session start, picks the active feature, works on it, updates evidence with verification artifacts, and transitions to `done`. The agent **cannot** put two features in `in-progress` simultaneously (WIP=1 invariant), and **cannot** mark `done` without populated evidence (Anti-Fabrication, see principle 02 Proof Loop).\n\nCombined with a minimal `init.sh` (canonical \"is the project healthy?\" check) and the existing `PROBLEMS.md` (incident log), these three artifacts form the standard harness for long-run projects.\n\nA sibling upstream rule, `rules\u002Flong-run-harness.md`, auto-triggers on these same artifacts and covers the same schema, WIP=1 invariant, and bootstrap procedure — a near-total duplicate of this module. It also names one genuinely non-duplicate practice, folded in below: a readiness gate to run before adopting this convention on a project, rather than skipping straight to the label.\n\n---\n\n## The Three-Artifact Harness\n\n| Artifact | Answers | When updated |\n|---|---|---|\n| `PROBLEMS.md` | What is **broken** right now? Recovery procedures? | When an incident happens or is resolved |\n| `feature_list.json` | What **features** exist? Which is active? What's done? | When status of a feature changes |\n| `init.sh` | Is the project **healthy** right now? (binary check) | When dependencies, tooling, or test commands change |\n\nThese are independent. A feature can be `blocked` because of a `PROBLEMS.md` entry. `init.sh` failing is its own kind of red — fix it before adding scope.\n\nThis sits **alongside** existing per-project conventions, not in place of them:\n\n- **AGENTS.md or project guidance \u002F AGENTS.md** — how to work here (rules and routing)\n- **`.hermes\u002Fhandoffs\u002F`** — tactical \"what to do next\" between sessions\n- **`.hermes\u002Fchronicles\u002F`** — strategic \"how we got here\" across months\n\nFeatures and incidents and health checks were the missing layer.\n\n---\n\n## feature_list.json — Schema\n\n```json\n{\n  \"features\": [\n    {\n      \"id\": \"feat-001\",\n      \"name\": \"User authentication\",\n      \"description\": \"Email + password login with JWT session tokens\",\n      \"dependencies\": [],\n      \"status\": \"done\",\n      \"evidence\": \"L1: tsc clean (commit a3f2c1); L2: pytest 12\u002F12 passed; L3: manual login flow verified in staging\"\n    }\n  ]\n}\n```\n\nIf a project also adopts the installed `feature-layer-architecture` skill's\n`docs\u002Flayers\u002F` tree, its `feature-new` skill appends entries here with three\nadditional fields — `layer`, `doc`, `branch` — using this same base schema and the\nsame `feat-NNN` id format (not a second, incompatible file). A tool that only reads\nthe six base fields above still works correctly. See `feature-new` for the exact\nreconciled shape.\n\n### Four states\n\n- `not-started` — defined, not yet picked up\n- `in-progress` — active work right now (WIP=1: at most **one** feature in this state)\n- `blocked` — cannot proceed, blocker named in `evidence`\n- `done` — all three validation layers passed with durable artifacts in `evidence`\n\n### Transition rules\n\n```\nnot-started  →  in-progress     # if all dependencies are 'done' AND no other in-progress\nin-progress  →  done            # only when evidence has L1+L2+L3 artifacts\nin-progress  →  blocked         # name the blocker in evidence\nblocked      →  in-progress     # after unblock (and WIP=1 still holds)\ndone         →  anything else   # FORBIDDEN. Regression → new feat-NNN\n```\n\nThe `done → ?` prohibition is important. If a previously-done feature regresses, **do not** flip it back to `in-progress`. Create a new feature `feat-NNN` named \"fix regression in feat-MMM\". This preserves the audit trail and forces explicit acknowledgment that something broke after being verified.\n\n---\n\n## WIP=1 Invariant\n\nAt most one feature in `in-progress` at any time, across the entire `feature_list.json`. This is enforced by convention; a session-finish routine concept can verify it (see \"Mechanical enforcement\" below).\n\n### Why WIP=1\n\nThe natural tendency under context pressure is to start a second feature when the first hits friction. \"I'll just begin on B while A's tests run.\" Two days later both are half-done, neither is verified, and the agent (or human) cannot tell which assumptions belong to which feature. WIP=1 forces a clean answer: either finish the current feature, formally block it, or roll it back to `not-started`. No middle states.\n\n### What to do when context demands switching\n\n- Current feature is **technically blocked** (external dependency, decision needed from user): mark `blocked` with reason in `evidence`, then start a new feature.\n- **Priorities changed** (user redirected): roll current back to `not-started` (or `blocked` if partial work matters), note the pivot in session handoff, start new feature.\n- **Never** leave two in `in-progress` simultaneously.\n\nThis pairs with principle 02 (Proof Loop) — both forbid the agent from making fuzzy claims about completion. WIP=1 prevents the related anti-pattern of fuzzy claims about *what is being worked on*.\n\n---\n\n## Evidence Field — What Belongs There\n\nWhen transitioning to `done`, `evidence` must reference durable artifacts at three layers:\n\n| Layer | Proves | Examples |\n|---|---|---|\n| **L1 — Static** | Syntax valid, types check, lint clean | `tsc --noEmit` output, `ruff check` output, commit hash where lint became clean |\n| **L2 — Runtime** | Tests pass, app starts, critical paths work | Test runner output file, log capture, .\u002Finit.sh exit 0 |\n| **L3 — System** | End-to-end behavior, integration verified | Screenshot, curl response log, video, user flow capture |\n\nExample acceptable evidence:\n\n```\nL1: tsc clean (commit a3f2c1) + ruff check passed\nL2: pytest tests\u002Ftest_auth.py 12\u002F12 passed (.agent\u002Fevidence\u002Ftest-output-2026-05-10.txt)\nL3: manual login + reset flow in staging (.agent\u002Fevidence\u002Fauth-flow.png)\n```\n\nNot acceptable as evidence:\n\n- \"Works for me\" \u002F \"Tested manually, looks good\"\n- \"I ran the tests\" without a file path\n- Reference to a chat or session — those disappear, artifacts persist\n\nThe three-layer requirement is the same gate as principle 02's Proof Loop. The difference: Proof Loop is per-task; feature evidence accumulates across the project lifetime, surviving every context reset and session change.\n\n---\n\n## init.sh — The Health Check\n\nA single executable script in the project root with one job: exit 0 if the project is healthy enough to work on, exit non-zero otherwise.\n\n```bash\n#!\u002Fbin\u002Fbash\nset -e\n\n# 1. Dependencies\nnpm install\n\n# 2. L1 — Static checks\nnpm run check  # tsc --noEmit\n\n# 3. L2 — Tests\nnpm test\n\n# 4. Build (if applicable)\nnpm run build\n\necho \"=== Initialization Complete ===\"\n```\n\n### Constraints\n\n- **Idempotent** — running twice in a row must not break anything\n- **Non-interactive** — no prompts; if credentials are needed, they come from env vars\n- **Fast** — target \u003C3 minutes from fresh clone to green. If your setup is slower, split into `init.sh` (essentials) and `init-full.sh` (everything).\n- **Free** — no paid API calls. The script should be runnable in CI on every PR.\n- **Local** — no deploys. `init.sh` proves the code works on this machine; deploy is separate.\n\n### What this replaces\n\nBefore `init.sh` convention: every new session spends 10-15 minutes piecing together commands from README + handoff + chronicle. With `init.sh`: 3 minutes from `git clone` to working state. The Learn Harness Engineering course measures this as a 5x speedup, consistent with our experience.\n\n### Bootstrap rule\n\nIf `init.sh` fails on a fresh checkout, **fix the baseline first**. Do not add new features on top of a broken baseline. The first task in any session should be \"is `.\u002Finit.sh` green?\" — if not, the only acceptable next task is making it green.\n\n---\n\n## When to Apply\n\n**Good fit** for `feature_list.json` + `init.sh`:\n\n- Projects with >5 distinct features\n- Projects spanning >5 sessions of work\n- Multi-developer or multi-agent collaboration\n- Anything you'd describe as \"long-running\" or \"ongoing\"\n\n**Skip** for:\n\n- Short-term projects (1-2 sessions total)\n- Projects with \u003C5 features\n- Pure research \u002F exploration where scope is intentionally fluid\n- Utility scripts and one-offs\n\nFor skipped projects, free-form notes in handoffs are fine. The overhead of maintaining `feature_list.json` is only worth it when there are enough features and sessions for the structure to pay back.\n\n---\n\n## Readiness Gate\n\nBefore adopting this convention on a project — not after — confirm the following, rather than adding `feature_list.json` and `init.sh` to a project whose foundations are still shaky. Adapted from `mvp-agent-blueprint`'s first-release-checklist pattern, narrowed to the specific artifacts this module owns.\n\nArtifact-level:\n\n- [ ] One primary job-to-be-done is stated at the top of the project's guidance file.\n- [ ] `init.sh` exists and exits 0 on a clean checkout in under 3 minutes.\n- [ ] `feature_list.json` exists with at least 5 features, including one `done` with populated evidence.\n- [ ] WIP=1 holds — exactly 0 or 1 feature `in-progress`.\n- [ ] `PROBLEMS.md` exists (even empty) as the incident log.\n- [ ] `.gitignore` covers regenerable output, secrets, and heavy binaries.\n\nProcess-level:\n\n- [ ] Autonomy level is stated explicitly (answer-only \u002F draft-only \u002F approval-gated \u002F autonomous-within-policy).\n- [ ] High-risk actions are draft-only or approval-gated, with risk classes named (see `agent-harness-design`).\n- [ ] Step\u002Fcost\u002Ftime budgets are declared for any agents the project runs (see `agent-harness-design`).\n- [ ] Trust labels are applied to external content the project consumes (see `agent-harness-design`).\n\nKnowledge-level:\n\n- [ ] Project guidance stays a short map, not an encyclopedia — detail moves to rules or docs.\n- [ ] A session-handoff workflow is in place between sessions.\n- [ ] Validation signals are declared — what \"this feature works\" means concretely (a named test, probe, or check).\n\nSafety-level:\n\n- [ ] Secrets are out of git, out of the guidance file, and out of scripts; a push-time scan is configured.\n\nWork through this iteratively — close one item, check it, move to the next. If it is not fully green, keep working in the ordinary mode instead: a project can go a long time without needing this convention. Adopting the convention and finishing the checklist \"later\" inverts the gate into decoration — the label should follow real readiness, not substitute for it.\n\n---\n\n## Mechanical Enforcement (Optional)\n\nA session-finish routine concept can verify the invariants automatically:\n\n```python\n# scripts\u002Fa reviewed guard candidate (pseudocode)\ndef check_feature_list(repo_root):\n    fl_path = repo_root \u002F \"feature_list.json\"\n    if not fl_path.exists():\n        return  # project doesn't use this convention\n    data = json.loads(fl_path.read_text())\n    in_progress = [f for f in data[\"features\"] if f[\"status\"] == \"in-progress\"]\n    if len(in_progress) > 1:\n        raise BlockedSession(f\"WIP=1 violated: {len(in_progress)} features in progress\")\n    # If this session edited the file to set status=done, evidence must be populated\n    if session_set_done() and not session_done_features_have_evidence():\n        raise BlockedSession(\"done without evidence\")\n```\n\nThis is a defence-in-depth layer. The primary enforcement is culture: the agent reads `feature_list.json` at session start, picks up the in-progress feature, and updates it correctly. The hook catches drift.\n\n---\n\n## Bootstrapping an Existing Project\n\nFor projects that already have handoffs and a chronicle but no `feature_list.json`:\n\n1. Read the last 5-10 handoff files plus the project chronicle\n2. Extract completed work → list of features → mark all `status: \"done\"` with evidence pointing to commit hashes\n3. Identify the **single** currently-active thread → one feature `in-progress`\n4. Identify planned work → features `not-started` with declared dependencies\n5. Identify blocked work → features `blocked` with reason in evidence\n6. Commit both `feature_list.json` and `init.sh` in one changeset: `harness: bootstrap feature_list + init.sh`\n\nThis takes 30-60 minutes for a typical 3-month-old project. The payoff is that every subsequent session starts from a clean machine-readable state instead of reading paragraph-form handoffs.\n\n---\n\n## Anti-Patterns\n\n- **50+ entries** in feature_list.json — this is a backlog, not a working list. Move non-active items to `BACKLOG.md`.\n- **`init.sh` that downloads multi-GB models or trains** — exceeds 3-minute target. Split: `init.sh` for quick health, `setup.sh` for one-time heavy lifting.\n- **`init.sh` that installs unpinned heavy dependencies without a locked version\u002Findex** — supply-chain risk and environment mismatch; pin exact versions.\n- **`done` with vague evidence** — \"tests pass\" is not evidence. The file path of the test output is.\n- **Two in-progress** to avoid blocking — name the actual blocker, set `blocked` honestly.\n- **Editing `feature_list.json` to silently roll back `done` → `in-progress`** — this hides regressions. Create a new fix feature instead.\n\n---\n\n## Relationship to Other Principles\n\n- **02 Proof Loop** — Anti-Fabrication and evidence requirements apply to every feature transition. Evidence field references the same kind of durable artifacts.\n- **04 Deterministic Orchestration** — `feature_list.json` is the state file the relay pattern reads. WIP=1 is a deterministic constraint a hook can verify mechanically.\n- **07 Codified Context** — `feature_list.json` is the canonical structured handoff between sessions; it survives compaction.\n- **16 Project Chronicles** — chronicles capture **why** decisions were made. feature_list captures **what** is currently done. Both are needed for long-run projects.\n- **MVP Agent Blueprint** — this module's Readiness Gate adapts its first-release-checklist pattern to the specific artifacts `feature_list.json` \u002F `init.sh` \u002F `PROBLEMS.md` own.\n\n---\n\n## Source\n\nTemplates and conceptual framework adapted from [walkinglabs\u002Flearn-harness-engineering](https:\u002F\u002Fgithub.com\u002Fwalkinglabs\u002Flearn-harness-engineering), MIT-licensed, specifically:\n\n- Lectures 06, 07, 08, 09 in the course documentation\n- `skills\u002Fharness-creator\u002Ftemplates\u002Ffeature-list.json` (schema)\n- `skills\u002Fharness-creator\u002Ftemplates\u002Ffeature-list.schema.json` (JSON schema)\n- `skills\u002Fharness-creator\u002Ftemplates\u002Finit.sh` (bootstrap script)\n\nOur adaptation:\n\n- Added explicit 3-layer evidence requirement (L1\u002FL2\u002FL3) — integrated with our existing Proof Loop principle (02)\n- Added WIP=1 invariant as a hard rule with `blocked` escape valve\n- Added relationship to existing PROBLEMS.md \u002F handoffs \u002F chronicles layers (not in the original course)\n- Drop-in templates available in `templates\u002Flong-run-project\u002F`\n",{"name":5,"license":14,"version":23,"metadata":24,"description":7},"0.1.0",{"hermes_config_kit":25},{"adapter":26,"conversion":27,"source_path":28,"source_repo":29},"hermes-agent-config-kit","adapted","principles\u002F27-feature-tracking.md","AnastasiyaW\u002Fclaude-code-config","main","07d02526f424fc814f8350d6b0b728e81576603d","hermes\u002Fskills\u002Flong-run-feature-tracking",0,"https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot","2026-11-14 18:25:27",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill long-run-feature-tracking",{"tool":43,"command":44},"bunx","bunx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill long-run-feature-tracking",{"tool":46,"command":47},"pnpm","pnpm dlx skills add hermes-jarvis-bot\u002Fhermes-agent-config-kit --skill long-run-feature-tracking","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Flong-run-feature-tracking\u002FSKILL.md",true,[],[52,58,65,72,79,86],{"skillNo":53,"name":54,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":18},"01M78BHH6VNT0MKZXGMP2F50X1","owner-voice","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fowner-voice\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M78BHD34XJHAK6EMZ7J1M80X","humanize-russian","Редактировать русский текст для ясности и естественного звучания, сохраняя факты,\nтермины, источники и жанр. Использовать когда: убрать канцелярит, сделать текст\nживее, отредактировать русский черновик или статью перед публикацией.\nНе определяет авторство и не обещает обход детекторов. Для английского —\nhumanize-english; для структуры и доказательств — article-structure-review.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fhumanize-russian\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M78BH9D09HCA7ECZMPEEAE5Y","humanize-english","Edit English drafts for clear, natural language while preserving facts, uncertainty,\ntechnical meaning, citations and genre. Use when: humanize this text, remove AI-style\nfiller, improve English prose, or edit an English article before publication.\nNot an authorship detector or a promise to bypass one. For Russian use\nhumanize-russian; for thesis\u002Fevidence and structure use article-structure-review.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fhumanize-english\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M78BH44Q25YMA16MC6FKZ2WY","caveman","Ultra-compressed reply mode. Cuts output tokens ~65-75% by answering in terse \"caveman\" style while keeping full technical accuracy, and shapes what survives: action first, state restated every turn, concrete time estimates, visible wins. Intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says \"caveman\", \"talk like caveman\", \"use caveman\", \"less tokens\", \"be brief\", or Russian \"пещерный режим\", \"говори как пещерный\", \"кратко\", \"меньше токенов\", \"экономь токены\" — or invokes \u002Fcaveman. Preserves the user's language (Russian in → Russian caveman out). Off only: \"stop caveman\" \u002F \"normal mode\" \u002F \"обычный режим\". Do not use merely because a response should be concise, and do not use when the user requests normal prose, a polished document, or exact wording.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Fcaveman\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M78BGY0JTMTW3J3YE5RQTHCK","article-structure-review","Проверить структуру технической статьи после черновика: задача читателя, жанр,\nопора значимых утверждений на доказательства и честные ограничения.\nUse after the first draft, before publication. Исправлять структуру, когда это\nвходит в запрос; не создавать обязательный конвейер из других навыков.\nДля локальной языковой редактуры — humanize-english или humanize-russian.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fwriting\u002Farticle-structure-review\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M78BGRJYZGK1TNTCHWGY32W3","video-post-production","Video post-production rules: audio mastering, color, captions, platform export. Use when: 'add music', 'add voiceover', 'export for tiktok', 'add captions', 'color grade', 'audio levels', 'master audio', 'export settings', 'platform requirements'. Covers FFmpeg patterns, audio chain, subtitle standards, and platform-specific export configs. Do NOT use to generate or animate the video frames\u002Fscenes in code (use remotion-production-guide), to write the script (use video-narrative-arc), or to evaluate script quality (use script-evaluator); this is the finishing stage applied to an already-rendered video.","https:\u002F\u002Fgithub.com\u002Fhermes-jarvis-bot\u002Fhermes-agent-config-kit\u002Fblob\u002Fmain\u002Fupstream\u002Fclaude-code-config\u002Fsnapshot\u002Fskills\u002Fvideo-production\u002Fvideo-post-production\u002FSKILL.md",[],[]]