[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MPA7XKYK44SM2DKHFZJCTP":3,"skill-same:01M6MPA7XKYK44SM2DKHFZJCTP":72},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":28,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":48,"commitSha":49,"skillDirectory":50,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":53,"ownerHtmlUrl":54,"ownerBlogUrl":55,"ownerDescription":56,"updatedAt":57,"install":58,"resources":71},"01M6MPA7XKYK44SM2DKHFZJCTP","writing-plans",null,"Use after design approval to create a dated executable workflow file with bite-sized tasks, exact file paths, and loop\u002Fgate definitions.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fyimwoo\u002Fhotl-plugin\u002Fskills\u002Fwriting-plans\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42],"writing","writing plans","plans","design","workflow","tasks","create","file","loop","approval","use","gate","paths","with","2026-11-10 10:12:30",false,"---\nname: writing-plans\ndescription: Use after design approval to create a dated executable workflow file with bite-sized tasks, exact file paths, and loop\u002Fgate definitions.\n---\n\n# Writing HOTL Plans\n\n## Overview\n\nProduce a dated workflow file that `loop-execution` can execute. The canonical filename is `YYYY-MM-DD-\u003Cslug>-workflow.md`, where `\u003Cslug>` is a short kebab-case identity derived from the intent (e.g., `2026-04-22-add-rate-limiting-workflow.md`). Each step should be 2-5 minutes of work. Include loop conditions and gates from the design's governance contract.\n\nWhen the accepted design defines execution limits, include only the relevant optional portable budget fields (`max_total_attempts`, `max_agents`, `max_cost_usd`, `max_elapsed_minutes`) in workflow frontmatter. Do not invent numeric budgets. `policy_profile` is descriptive and never grants host permissions.\n\n**Announce:** \"I'm using the writing-plans skill to create the executable workflow.\"\n\n## Output Filename\n\nSave as `YYYY-MM-DD-\u003Cslug>-workflow.md`, where:\n\n- `YYYY-MM-DD` is the current local date\n- `\u003Cslug>` is a short kebab-case semantic identity derived from the intent (e.g., `add-user-auth`, `refactor-api`)\n\nExamples:\n\n- `2026-04-22-add-user-auth-workflow.md`\n- `2026-04-22-refactor-api-workflow.md`\n\nThe date makes workflow revisions easy to sort chronologically. The semantic identity remains `\u003Cslug>`, not the date-prefixed filename.\n\n## Semantic Identity\n\nThe canonical workflow filename is human-friendly, but the stable execution identity is still `\u003Cslug>`.\n\n- Filename: `YYYY-MM-DD-\u003Cslug>-workflow.md`\n- Semantic identity: `\u003Cslug>`\n- Default branch derivation: `hotl\u002F\u003Cslug>`\n\nWhen a later workflow revision is created for the same feature, it should use a new dated filename while keeping the same semantic identity.\n\n## Output Directory\n\nDefault: `docs\u002Fplans`. Opt-in override via `.hotl\u002Fconfig.yml: workflows_dir: \u003Cpath>`. Resolution procedure:\n\n1. Resolve the install path of `hotl-config-resolve.sh` using the same six-location rule documented for `document-lint.sh` and `hotl-config.sh` (see `skills\u002Fdocument-review\u002FSKILL.md`):\n   1. In-repo: `scripts\u002Fhotl-config-resolve.sh`\n   2. Codex native-skills install: `~\u002F.codex\u002Fhotl\u002Fscripts\u002Fhotl-config-resolve.sh`\n   3. Codex plugin install: `~\u002F.codex\u002Fplugins\u002Fhotl-source\u002Fscripts\u002Fhotl-config-resolve.sh`\n   4. Codex plugin cache fallback: `~\u002F.codex\u002Fplugins\u002Fcache\u002Fcodex-plugins\u002Fhotl\u002F*\u002Fscripts\u002Fhotl-config-resolve.sh`\n   5. Cline install fallback: `~\u002F.cline\u002Fhotl\u002Fscripts\u002Fhotl-config-resolve.sh`\n   6. Claude Code plugin fallback: `~\u002F.claude\u002Fplugins\u002Fhotl\u002Fscripts\u002Fhotl-config-resolve.sh`\n\n2. Invoke the resolver as a command proxy — it forwards argv to `hotl-config.sh`, no intermediate path-locator step:\n\n   ```bash\n   bash \u003Cresolved-hotl-config-resolve.sh> get workflows_dir --default=docs\u002Fplans\n   ```\n\n3. Use the returned directory as the output location. `docs\u002Fplans` is the canonical default. Any other value is an opt-in override — write `YYYY-MM-DD-\u003Cslug>-workflow.md` inside that directory, creating it if needed.\n\nProjects with no `.hotl\u002Fconfig.yml` receive `docs\u002Fplans` from the `--default=docs\u002Fplans` fallback.\n\nFormat:\n\n```markdown\n---\nintent: [from design's intent contract]\nsuccess_criteria: [from design's intent contract]\nrisk_level: low | medium | high\nauto_approve: true | false\n# branch: custom\u002Fbranch-name   # optional — execution derives hotl\u002F\u003Cslug> if absent\n# worktree: host                # optional: only when a host tool already put this task on a feature-branch worktree\n# dirty_worktree: allow         # optional — proceed even if non-HOTL files are uncommitted\n---\n\n## Steps\n\n- [ ] **Step N: [Step name]**\naction: [what to do]\nloop: false | until [condition]\nmax_iterations: [number, default 3]\nverify: [scalar command OR typed block]\ngate: human | auto   # optional\n```\n\n**CRITICAL — field indentation:** `action:`, `loop:`, `verify:`, `max_iterations:`, and `gate:` MUST start at column 0 (no leading spaces). The document linter matches `^action:`, `^loop:`, etc. — any indentation (even 2 spaces under the list item) will fail validation. Only the sub-fields of structured `verify:` blocks (like `type:`, `path:`, `assert:`) are indented.\n\n## Typed Verification\n\nChoose the appropriate verify type for each step:\n\n- **shell** — for test suites, linters, build commands (default; scalar shorthand accepted)\n- **browser** — for UI work requiring visual inspection (capability-gated; falls back to human-review)\n- **human-review** — for subjective quality checks with no automated signal\n- **artifact** — for verifying files\u002Foutputs exist and meet criteria\n\n```yaml\n# Scalar shorthand (type: shell)\nverify: pytest tests\u002F -v\n\n# Structured\nverify:\n  type: browser\n  url: http:\u002F\u002Flocalhost:3000\u002Fdashboard\n  check: priority badge renders with correct color\n\n# Artifact with structured assert\nverify:\n  type: artifact\n  path: migrations\n  assert:\n    kind: matches-glob\n    value: \"*.sql\"\n\n# Greenfield scaffold check\nverify:\n  type: artifact\n  path: src\n  assert:\n    kind: exists\n\n# Multiple checks per step\nverify:\n  - type: shell\n    command: npm test\n  - type: artifact\n    path: coverage\u002Flcov.info\n    assert:\n      kind: exists\n```\n\n## Step Granularity\n\nBreak work into atomic steps:\n- \"Write failing test for X\" (loop: false, verify: pytest)\n- \"Implement X\" (loop: until tests pass, verify: pytest)\n- \"Fix lint errors\" (loop: until clean, verify: ruff check .)\n- \"Verify UI renders correctly\" (loop: false, verify: type: browser)\n- \"Human review of security logic\" (loop: false, gate: human — REQUIRED for risk_level: high)\n\n## No Placeholders\n\nEvery workflow step must contain enough concrete information for an agent to execute it without guessing. These are plan failures:\n\n- `TBD`, `TODO`, `later`, `fill in`, or empty sections\n- Vague instructions such as \"add validation\", \"handle edge cases\", \"improve error handling\", or \"update docs\" without exact files and success signals\n- \"Similar to Step N\", \"repeat the pattern\", or references that require reading another step to know what to do\n- Functions, files, commands, artifacts, or config keys referenced before they are defined\n- Verification that does not prove the step's action worked\n- Human-review checks where an automated shell, browser, or artifact check would provide a deterministic signal\n\nIf a step cannot be made concrete yet, stop and ask for clarification instead of saving the workflow.\n\n## Branch And Worktree Authoring Guidance\n\n- Default execution branch is `hotl\u002F\u003Cslug>` unless the workflow frontmatter sets `branch: ...`\n- Default execution mode is an isolated worktree; HOTL copies the workflow into that worktree at the same relative path before execution\n- Use `branch:` only when downstream tooling or verify logic truly depends on a specific branch name\n- Use `worktree: false` only when execution must stay in the current checkout rather than a separate worktree\n- To keep execution on the exact current branch, set both `branch: \u003Ccurrent-branch>` and `worktree: false`\n- Use `worktree: host` when Codex or another host tool already created the execution worktree and HOTL should use the current feature branch exactly as-is\n- Avoid brittle verify steps such as `git branch --show-current | grep '^feature\u002F'` unless the workflow pins `branch:` to match that exact convention\n- If a step must confirm the workflow file exists, use its repo-relative path; HOTL preserves that relative path inside the isolated worktree\n- If the workflow is authored on a non-`main`\u002F`master` branch and does not pin `branch:` or `worktree:`, execution should pause and ask whether to continue on the current branch or use HOTL's isolated execution branch\u002Fworktree\n\nExample:\n\n- If verify expects `pv6-ui\u002F...`, set `branch: pv6-ui\u002Fplan-amendments`\n- If branch name does not matter, let HOTL derive `hotl\u002F\u003Cslug>` and do not assert a custom prefix\n- If you want to keep using `pv6-ui\u002Fplan-amendments` itself, set `branch: pv6-ui\u002Fplan-amendments` and `worktree: false`\n- If Codex already opened the task in a worktree on `pv6-ui\u002Fplan-amendments`, set `worktree: host` and omit `branch:`\n\n## Artifact Verification Rules\n\n- Prefer `kind: exists` when the step creates a new file or directory from scratch\n- Use `kind: matches-glob` only when `path` is an existing directory and `value` is a filename glob such as `*.tsx` or `*.md`\n- Do not put directory segments in `value`; write `path: src` with `value: \"*.tsx\"`, not `path: .` with `value: \"src\u002F*.tsx\"`\n- For greenfield frontend scaffolds, Step 1 should usually verify `src` or `package.json` with `kind: exists`, not `matches-glob`\n\n## risk_level Guidelines\n\n- **low:** UI changes, new endpoints, non-critical features\n- **medium:** Schema changes, refactors, performance work\n- **high:** Auth\u002Fauthz, encryption, privacy logic, billing, multi-tenant isolation\n\n`risk_level: high` **always** generates `gate: human` on security-sensitive steps, regardless of `auto_approve`.\n\n## Self-Check Loop\n\nAfter saving the workflow file, run a self-check before offering execution options. Review the workflow for:\n\n- **Step sizing** — each step should be 2-5 minutes of atomic work\n- **No placeholders** — no TBD\u002FTODO text, vague instructions, undefined references, or \"similar to Step N\" shortcuts\n- **Verify coverage** — every looped step has a verify command that tests what the step claims\n- **Gate placement** — risky steps (auth, encryption, billing, secrets) have `gate: human`\n- **Loop safety** — `max_iterations` is reasonable (typically 3-5)\n- **Ordering** — logical dependencies between steps are respected\n\nIf issues are found, fix them in the workflow file and re-check until clean. Do not ask the user to review — this is an internal quality pass.\n\n## After Saving\n\nOnce the self-check passes, offer execution options:\n\n**\"Workflow saved to `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`. How would you like to execute?\"**\n\nPresent the three execution modes using the current host tool's native invocation style. Never show Claude Code slash commands in Codex or any other skill-based agent.\n\nUse these mappings:\n\n1. **Loop execution (this session)** — runs steps autonomously with auto-approve\n   - **Codex:** ask me to use `$hotl:loop-execution` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n   - **Claude Code:** `\u002Fhotl:loop \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n2. **Manual execution** — linear execution with explicit checkpoints\n   - **Codex:** ask me to use `$hotl:executing-plans` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n   - **Claude Code:** `\u002Fhotl:execute-plan \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n3. **Subagent execution (this session)** — delegates implementation-friendly steps to fresh subagents while the controller keeps gates and verification\n   - **Codex:** ask me to use `$hotl:subagent-execution` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n   - **Claude Code:** `\u002Fhotl:subagent-execute \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n\nIf a previous run was interrupted, point the user to the host tool's native resume entry point.\n- **Codex:** ask me to use `$hotl:resuming` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n- **Claude Code:** `\u002Fhotl:resume \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n- **Other agents:** use that agent's native skill\u002Fcommand invocation instead of inventing Claude-style slash commands\n\n*(Always tell the user the exact workflow filename so they can pass it to the execution request if multiple workflow files exist.)*\n\nIf execution starts from a non-`main`\u002F`master` branch and the workflow does not already pin `branch:` or `worktree:`, tell the user HOTL will ask one more continuity question at execution time:\n- continue on the current branch in this checkout\n- use HOTL's isolated execution branch\u002Fworktree (recommended)\n- choose a custom execution branch\n","# Writing HOTL Plans\n\n## Overview\n\nProduce a dated workflow file that `loop-execution` can execute. The canonical filename is `YYYY-MM-DD-\u003Cslug>-workflow.md`, where `\u003Cslug>` is a short kebab-case identity derived from the intent (e.g., `2026-04-22-add-rate-limiting-workflow.md`). Each step should be 2-5 minutes of work. Include loop conditions and gates from the design's governance contract.\n\nWhen the accepted design defines execution limits, include only the relevant optional portable budget fields (`max_total_attempts`, `max_agents`, `max_cost_usd`, `max_elapsed_minutes`) in workflow frontmatter. Do not invent numeric budgets. `policy_profile` is descriptive and never grants host permissions.\n\n**Announce:** \"I'm using the writing-plans skill to create the executable workflow.\"\n\n## Output Filename\n\nSave as `YYYY-MM-DD-\u003Cslug>-workflow.md`, where:\n\n- `YYYY-MM-DD` is the current local date\n- `\u003Cslug>` is a short kebab-case semantic identity derived from the intent (e.g., `add-user-auth`, `refactor-api`)\n\nExamples:\n\n- `2026-04-22-add-user-auth-workflow.md`\n- `2026-04-22-refactor-api-workflow.md`\n\nThe date makes workflow revisions easy to sort chronologically. The semantic identity remains `\u003Cslug>`, not the date-prefixed filename.\n\n## Semantic Identity\n\nThe canonical workflow filename is human-friendly, but the stable execution identity is still `\u003Cslug>`.\n\n- Filename: `YYYY-MM-DD-\u003Cslug>-workflow.md`\n- Semantic identity: `\u003Cslug>`\n- Default branch derivation: `hotl\u002F\u003Cslug>`\n\nWhen a later workflow revision is created for the same feature, it should use a new dated filename while keeping the same semantic identity.\n\n## Output Directory\n\nDefault: `docs\u002Fplans`. Opt-in override via `.hotl\u002Fconfig.yml: workflows_dir: \u003Cpath>`. Resolution procedure:\n\n1. Resolve the install path of `hotl-config-resolve.sh` using the same six-location rule documented for `document-lint.sh` and `hotl-config.sh` (see `skills\u002Fdocument-review\u002FSKILL.md`):\n   1. In-repo: `scripts\u002Fhotl-config-resolve.sh`\n   2. Codex native-skills install: `~\u002F.codex\u002Fhotl\u002Fscripts\u002Fhotl-config-resolve.sh`\n   3. Codex plugin install: `~\u002F.codex\u002Fplugins\u002Fhotl-source\u002Fscripts\u002Fhotl-config-resolve.sh`\n   4. Codex plugin cache fallback: `~\u002F.codex\u002Fplugins\u002Fcache\u002Fcodex-plugins\u002Fhotl\u002F*\u002Fscripts\u002Fhotl-config-resolve.sh`\n   5. Cline install fallback: `~\u002F.cline\u002Fhotl\u002Fscripts\u002Fhotl-config-resolve.sh`\n   6. Claude Code plugin fallback: `~\u002F.claude\u002Fplugins\u002Fhotl\u002Fscripts\u002Fhotl-config-resolve.sh`\n\n2. Invoke the resolver as a command proxy — it forwards argv to `hotl-config.sh`, no intermediate path-locator step:\n\n   ```bash\n   bash \u003Cresolved-hotl-config-resolve.sh> get workflows_dir --default=docs\u002Fplans\n   ```\n\n3. Use the returned directory as the output location. `docs\u002Fplans` is the canonical default. Any other value is an opt-in override — write `YYYY-MM-DD-\u003Cslug>-workflow.md` inside that directory, creating it if needed.\n\nProjects with no `.hotl\u002Fconfig.yml` receive `docs\u002Fplans` from the `--default=docs\u002Fplans` fallback.\n\nFormat:\n\n```markdown\n---\nintent: [from design's intent contract]\nsuccess_criteria: [from design's intent contract]\nrisk_level: low | medium | high\nauto_approve: true | false\n# branch: custom\u002Fbranch-name   # optional — execution derives hotl\u002F\u003Cslug> if absent\n# worktree: host                # optional: only when a host tool already put this task on a feature-branch worktree\n# dirty_worktree: allow         # optional — proceed even if non-HOTL files are uncommitted\n---\n\n## Steps\n\n- [ ] **Step N: [Step name]**\naction: [what to do]\nloop: false | until [condition]\nmax_iterations: [number, default 3]\nverify: [scalar command OR typed block]\ngate: human | auto   # optional\n```\n\n**CRITICAL — field indentation:** `action:`, `loop:`, `verify:`, `max_iterations:`, and `gate:` MUST start at column 0 (no leading spaces). The document linter matches `^action:`, `^loop:`, etc. — any indentation (even 2 spaces under the list item) will fail validation. Only the sub-fields of structured `verify:` blocks (like `type:`, `path:`, `assert:`) are indented.\n\n## Typed Verification\n\nChoose the appropriate verify type for each step:\n\n- **shell** — for test suites, linters, build commands (default; scalar shorthand accepted)\n- **browser** — for UI work requiring visual inspection (capability-gated; falls back to human-review)\n- **human-review** — for subjective quality checks with no automated signal\n- **artifact** — for verifying files\u002Foutputs exist and meet criteria\n\n```yaml\n# Scalar shorthand (type: shell)\nverify: pytest tests\u002F -v\n\n# Structured\nverify:\n  type: browser\n  url: http:\u002F\u002Flocalhost:3000\u002Fdashboard\n  check: priority badge renders with correct color\n\n# Artifact with structured assert\nverify:\n  type: artifact\n  path: migrations\n  assert:\n    kind: matches-glob\n    value: \"*.sql\"\n\n# Greenfield scaffold check\nverify:\n  type: artifact\n  path: src\n  assert:\n    kind: exists\n\n# Multiple checks per step\nverify:\n  - type: shell\n    command: npm test\n  - type: artifact\n    path: coverage\u002Flcov.info\n    assert:\n      kind: exists\n```\n\n## Step Granularity\n\nBreak work into atomic steps:\n- \"Write failing test for X\" (loop: false, verify: pytest)\n- \"Implement X\" (loop: until tests pass, verify: pytest)\n- \"Fix lint errors\" (loop: until clean, verify: ruff check .)\n- \"Verify UI renders correctly\" (loop: false, verify: type: browser)\n- \"Human review of security logic\" (loop: false, gate: human — REQUIRED for risk_level: high)\n\n## No Placeholders\n\nEvery workflow step must contain enough concrete information for an agent to execute it without guessing. These are plan failures:\n\n- `TBD`, `TODO`, `later`, `fill in`, or empty sections\n- Vague instructions such as \"add validation\", \"handle edge cases\", \"improve error handling\", or \"update docs\" without exact files and success signals\n- \"Similar to Step N\", \"repeat the pattern\", or references that require reading another step to know what to do\n- Functions, files, commands, artifacts, or config keys referenced before they are defined\n- Verification that does not prove the step's action worked\n- Human-review checks where an automated shell, browser, or artifact check would provide a deterministic signal\n\nIf a step cannot be made concrete yet, stop and ask for clarification instead of saving the workflow.\n\n## Branch And Worktree Authoring Guidance\n\n- Default execution branch is `hotl\u002F\u003Cslug>` unless the workflow frontmatter sets `branch: ...`\n- Default execution mode is an isolated worktree; HOTL copies the workflow into that worktree at the same relative path before execution\n- Use `branch:` only when downstream tooling or verify logic truly depends on a specific branch name\n- Use `worktree: false` only when execution must stay in the current checkout rather than a separate worktree\n- To keep execution on the exact current branch, set both `branch: \u003Ccurrent-branch>` and `worktree: false`\n- Use `worktree: host` when Codex or another host tool already created the execution worktree and HOTL should use the current feature branch exactly as-is\n- Avoid brittle verify steps such as `git branch --show-current | grep '^feature\u002F'` unless the workflow pins `branch:` to match that exact convention\n- If a step must confirm the workflow file exists, use its repo-relative path; HOTL preserves that relative path inside the isolated worktree\n- If the workflow is authored on a non-`main`\u002F`master` branch and does not pin `branch:` or `worktree:`, execution should pause and ask whether to continue on the current branch or use HOTL's isolated execution branch\u002Fworktree\n\nExample:\n\n- If verify expects `pv6-ui\u002F...`, set `branch: pv6-ui\u002Fplan-amendments`\n- If branch name does not matter, let HOTL derive `hotl\u002F\u003Cslug>` and do not assert a custom prefix\n- If you want to keep using `pv6-ui\u002Fplan-amendments` itself, set `branch: pv6-ui\u002Fplan-amendments` and `worktree: false`\n- If Codex already opened the task in a worktree on `pv6-ui\u002Fplan-amendments`, set `worktree: host` and omit `branch:`\n\n## Artifact Verification Rules\n\n- Prefer `kind: exists` when the step creates a new file or directory from scratch\n- Use `kind: matches-glob` only when `path` is an existing directory and `value` is a filename glob such as `*.tsx` or `*.md`\n- Do not put directory segments in `value`; write `path: src` with `value: \"*.tsx\"`, not `path: .` with `value: \"src\u002F*.tsx\"`\n- For greenfield frontend scaffolds, Step 1 should usually verify `src` or `package.json` with `kind: exists`, not `matches-glob`\n\n## risk_level Guidelines\n\n- **low:** UI changes, new endpoints, non-critical features\n- **medium:** Schema changes, refactors, performance work\n- **high:** Auth\u002Fauthz, encryption, privacy logic, billing, multi-tenant isolation\n\n`risk_level: high` **always** generates `gate: human` on security-sensitive steps, regardless of `auto_approve`.\n\n## Self-Check Loop\n\nAfter saving the workflow file, run a self-check before offering execution options. Review the workflow for:\n\n- **Step sizing** — each step should be 2-5 minutes of atomic work\n- **No placeholders** — no TBD\u002FTODO text, vague instructions, undefined references, or \"similar to Step N\" shortcuts\n- **Verify coverage** — every looped step has a verify command that tests what the step claims\n- **Gate placement** — risky steps (auth, encryption, billing, secrets) have `gate: human`\n- **Loop safety** — `max_iterations` is reasonable (typically 3-5)\n- **Ordering** — logical dependencies between steps are respected\n\nIf issues are found, fix them in the workflow file and re-check until clean. Do not ask the user to review — this is an internal quality pass.\n\n## After Saving\n\nOnce the self-check passes, offer execution options:\n\n**\"Workflow saved to `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`. How would you like to execute?\"**\n\nPresent the three execution modes using the current host tool's native invocation style. Never show Claude Code slash commands in Codex or any other skill-based agent.\n\nUse these mappings:\n\n1. **Loop execution (this session)** — runs steps autonomously with auto-approve\n   - **Codex:** ask me to use `$hotl:loop-execution` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n   - **Claude Code:** `\u002Fhotl:loop \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n2. **Manual execution** — linear execution with explicit checkpoints\n   - **Codex:** ask me to use `$hotl:executing-plans` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n   - **Claude Code:** `\u002Fhotl:execute-plan \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n3. **Subagent execution (this session)** — delegates implementation-friendly steps to fresh subagents while the controller keeps gates and verification\n   - **Codex:** ask me to use `$hotl:subagent-execution` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n   - **Claude Code:** `\u002Fhotl:subagent-execute \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n\nIf a previous run was interrupted, point the user to the host tool's native resume entry point.\n- **Codex:** ask me to use `$hotl:resuming` on `\u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n- **Claude Code:** `\u002Fhotl:resume \u003Cresolved-workflows-dir>\u002FYYYY-MM-DD-\u003Cslug>-workflow.md`\n- **Other agents:** use that agent's native skill\u002Fcommand invocation instead of inventing Claude-style slash commands\n\n*(Always tell the user the exact workflow filename so they can pass it to the execution request if multiple workflow files exist.)*\n\nIf execution starts from a non-`main`\u002F`master` branch and the workflow does not already pin `branch:` or `worktree:`, tell the user HOTL will ask one more continuity question at execution time:\n- continue on the current branch in this checkout\n- use HOTL's isolated execution branch\u002Fworktree (recommended)\n- choose a custom execution branch\n",{"name":5,"description":7},"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fyimwoo\u002Fhotl-plugin\u002Fskills\u002Fwriting-plans",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill writing-plans",{"tool":64,"command":65},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill writing-plans",{"tool":67,"command":68},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill writing-plans","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fyimwoo\u002Fhotl-plugin\u002Fskills\u002Fwriting-plans\u002FSKILL.md",true,[],[73,83,91,99,107,115],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":78,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":82},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":87,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":82},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":95,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":82},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":103,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":82},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":111,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":82},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":119,"sourceHtmlUrl":120,"categories":121,"keywords":122,"publishedAt":82},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]