[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CF05ZQHH9X8BFDGE4JW0JS":3,"skill-same:01M7CF05ZQHH9X8BFDGE4JW0JS":45},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"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":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":28,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M7CF05ZQHH9X8BFDGE4JW0JS","rr-build-all",null,"vanssata\u002Fclaude-agentic-sdlc-module","vanssata","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1220804?v=4","01M7CEZV36E48TC48Q0V69PFE0","Ivan kakurov","01M7CEZV3SVZE2WWZYMDW13X2X",3,"MIT","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Fskills\u002Frr-build-all\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: rr-build-all\ndescription: Run a whole programme plan as one task in a repository that has .ai\u002F — every row of docs\u002Fsdlc\u002Fplans\u002F\u003Cslug>.md becomes a step with its files, tests and dependencies; the ready steps go to parallel implementer subagents on the BALANCED tier, serialized where one depends on another; after the last step the suite runs once to the end, every failure is recorded and fixed by parallel fixers as one batch, then one review at the mode's tier, its findings fixed by parallel fixers as one batch, then tests and the review again until it passes; the pass past the batch budget (three failed reviews) is the human's. Use on \"\u002Frr-build-all\", \"run the whole plan\", \"all points in one task\", \"sdlc-lite\", \"batch build\".\nargument-hint: [\u003Cpath to docs\u002Fsdlc\u002Fplans\u002F\u003Cslug>.md>] | --resume\n---\n\n# \u002Frr-build-all $ARGUMENTS\n\nOne task for the whole plan — the loop a developer otherwise runs by hand: implement every\nrow, test once, fix everything, review, fix everything, again, until the review passes. The\nmode (light, guarded, full) is still the weight's; batch only changes who implements, how\nfailures are fixed, and how many rounds run before the human. Resolve the install root once:\n\n```bash\nfor AI_HOME in \"${CLAUDE_CONFIG_DIR:-$HOME\u002F.claude}\" \"${CODEX_HOME:-$HOME\u002F.codex}\"; do\n  [ -d \"$AI_HOME\u002Fskills\u002Frr-build-all\" ] && break\ndone\nSTATE=\"python3 $AI_HOME\u002Fskills\u002Fai-task\u002Fstate.py\"\n```\n\nShared files resolve through `\"$AI_HOME\u002Friskrails\u002Fai-path\" \u003Cpath>`. Read `policies\u002Fsafety.md`\nand `using-riskrails.md` once per session. Every rule of `\u002Frr-build` holds here: nothing outside\nthe task's files, no commit, a refusal names the way out and is never reworded around.\n\n## 1. Resume or start\n\n```bash\n$STATE handoff --print\n```\n\n- **A task in flight whose handoff says `Batch: on`:** continue at its stage — §3 while\n  `steps --ready` prints a step, §4 after the last `step-done`, §5 after a green verify.\n- **A task in flight without that line:** it runs the standard way — say so and hand over to\n  `\u002Frr-build`. Only the human turns the loop on mid-task (`batch on`, §Switching).\n- **No task in flight:** §2.\n\n## 2. The plan → one task\n\n`$ARGUMENTS` is a programme plan (default: the newest `docs\u002Fsdlc\u002Fplans\u002F*.md`). Its spec's\n**Flagged concerns** are resolved or accepted, as `\u002Frr-plan` demands; a plan with an open\nblocking question is not started.\n\n1. **Rows → steps.** Every row of **Order of work** that an agent does becomes one step\n   `{\"id\", \"files\", \"tests\", \"after\"}`: the row's files, the tests it names, and in `after` the\n   rows it depends on — its *Consumes* line, a stated dependency, or an earlier row it builds\n   on. A refactoring row stays its own task (C3): leave it out and say so. A human row (G1, …)\n   is not a step: every row after it names in `after` every step before it, so nothing past\n   the human row is ready until the rows before it are done — and its question (3.) answered.\n2. **Weigh and start** — `\u002Frr-plan`'s \"A task\" section, on the union of the steps' files, then:\n\n   ```bash\n   $STATE start --goal \"\u003Cthe plan's title>\" --workflow feature --files \"\u003Cevery file of every step>\" \\\n     --tests \"\u003Cevery test>\" [--declare …] --plan-file \u002Fpath\u002Fsteps.json \\\n     --batch --by \"\u003Cthe human's name>\" --why \"\u002Frr-build-all docs\u002Fsdlc\u002Fplans\u002F\u003Cslug>.md\" \\\n     [--mode \u003Chigher> --by \"\u003Cname>\"]\n   ```\n\n   An `after` id must exist and form no cycle; two steps that name one file are ordered by\n   `start` itself (`ordered: s4 after s1 (shares …)`), so two implementers never hold one file.\n   In full mode the written plan is the programme plan the human already accepted. With\n   `batch.default` true in `.ai\u002Fpolicies\u002Fweight.json` the flag is implied; `--no-batch` opts out.\n3. **A human row** becomes a question the moment every step before it is done — asked\n   **before** the next wave is dispatched:\n   `$STATE ask \"G1 is yours: \u003Cwhat>\" --option \"A: done\" --option \"B: not yet\" --recommend A`,\n   then **stop**. Never dispatch while a question is pending (`step-done` and `close` exit 4;\n   `steps --ready` still answers); `\u002Frr-build-all --resume` goes on once the human answered.\n\n## 3. Implement, by waves\n\n```bash\n$STATE steps --ready               # the next wave: the steps whose dependencies are done, one id per line\n$STATE steps --json                # batch_policy.tier and .parallel: the implementers' tier, how many at once\n$STATE profile --tier \u003Cthat tier>  # the tier's model in this runtime (BALANCED as shipped)\n```\n\n**Resuming mid-wave** (after a compaction or a stop): before dispatching, compare\n`git status --short` with each ready step's files. A ready step whose files already changed\nwas dispatched before and never recorded — do not send it again blind: brief its implementer\nwith that partial diff to finish, or `step-done` it if its tests pass.\n\nDispatch one subagent per ready step, at most `batch.parallel` (shipped 3) at once — under\nClaude Code a `general-purpose` agent with `model:` set to that tier's model; under Codex a\nspawn with no agent named, on that model. Its brief is the implementer contract,\n`agents\u002Fimplementer.md` in the shared tree, filled in:\n\n- **Goal** — the row's text from the plan. **Allowed files** — the step's files, nothing else.\n  **Forbidden files** — every other step's files, \"another implementer holds them\".\n  **Required behaviour \u002F behaviour that must not change** — from the row and the spec.\n  **Required tests** — the step's tests.\n- Its own tests, run directly: `step_test_command` from `.ai\u002Fpolicies\u002Ftesting.md` with the\n  step's tests in place of `{files}`. **A subagent never calls `state.py`** — one state file,\n  one writer, this session.\n- It returns `## RESULT`, or `## SCOPE_CHANGE_REQUIRED`, or `## QUESTIONS_NEEDED`.\n\nAfter the wave, in this session:\n\n- `RESULT` → `$STATE step-done \u003Cid>` for each step. Exit 6 names a file outside the task:\n  `$STATE start --amend --files \"+\u003Cpath>\" --why \"…\"`, then `step-done` again.\n- `SCOPE_CHANGE_REQUIRED` → amend as above and send the same step again. A file that belongs\n  to another change goes into `.ai\u002Fproject\u002Fknown-risks.md`, not into the task.\n- `QUESTIONS_NEEDED` → `$STATE ask --batch \u002Fpath\u002Fquestions.json`, show the questions, **stop**;\n  the human answers, and `\u002Frr-build-all --resume` goes on from the handoff.\n\nThen `steps --ready` again. Nothing ready and nothing pending: §4. Keep this session small:\nread `## RESULT` blocks, never a subagent's diff; `git status --short` is the only view of the\ntree you need between waves.\n\n## 4. Tests once, failures fixed as one parallel batch\n\nAfter the last wave, `$STATE step-done \u003Cid> && $STATE verify --scope all`:\n\n```bash\n$STATE verify --scope all\n$STATE verify --scope all --e2e not-applicable \"\u003Cwhy the change reaches no e2e flow>\"\n```\n\nGreen: §5. **Red:** a FAST `log-reader` reads the log by path and returns every failure\nclassified — new regression, existing failure, environment, unknown (`testing.md`, \"Reading a\nfailure\") — each with `file:line`. Then:\n\n- an existing failure → one entry in `.ai\u002Fproject\u002Fknown-risks.md`; an environment failure →\n  one more run, no code change;\n- the new regressions, grouped by the file they point at → one fixer subagent per group, in\n  parallel, at most `batch.parallel`, on the batch tier, briefed with the failures' excerpts\n  and the files it may touch (that group's files, inside the task's). Fix the code, never the\n  test; report `files_changed` and what stayed unfixed;\n- one run: `$STATE verify --scope all --after-fix \"\u003Cn> regressions: \u003Cone line each>\"`.\n\nRed again → the same once more. Exit 6 `SUITE_BUDGET` — past `batch.max_suite_runs`\n(shipped 8) — is the human's run: the pick in the dialog, or the command it prints, in their\nown terminal. **Stop.**\n\n## 5. Review, findings fixed as one parallel batch, again until it passes\n\nOne `ai-reviewer` at the mode's tier, briefed as `\u002Frr-review` §2 says: the whole diff since the\ntask's base, the task's files, the tests, the last verify log; unbounded \"look for\" in full\nmode, the security focus when the domain axis is 2. Record it verbatim:\n\n```bash\n$STATE review --verdict \u003Cpass|pass_with_findings|blockers_open> --tier \u003CBALANCED|STRONG> \\\n  [--focus security] --report - \u003C\u003C'EOF'\n\u003Cthe reviewer's FINDINGS and VERDICT, unchanged>\nEOF\n```\n\n- **pass** → §6.\n- **pass_with_findings** → the in-scope HIGHs, grouped by file, one fixer per group in\n  parallel; `$STATE verify --scope all --after-fix \"H1,H2: …\"`; `$STATE review --closes \"H1,H2\"`\n  → §6. MEDIUM and LOW: `memory risk add` as `\u002Frr-review` §4 says, for every verdict; only a\n  BLOCKER, a HIGH or a regression buys an `--after-fix` run.\n- **blockers_open** → every in-scope BLOCKER and HIGH, grouped by file, one fixer per group in\n  parallel; `$STATE verify --scope all --after-fix \"B1,H1: …\"` (red: §4's batch); then\n  **§5 again, a full pass** — a fix batch across many files is a new change, not a scoped one.\n\nExit 6 `REVIEW_BUDGET` after `batch.review_rounds_max` passes (shipped 3 — three failed\nreviews) — the next verdict is the human's: the pick in the dialog, or the command it prints,\nin their own terminal. Show the open findings and **stop**. A `verify` past its budget inside\nthis loop is `SUITE_BUDGET`, handled the same way.\n\n## 6. Ship\n\n`\u002Frr-ship` — `close` measures the two facts and, in full mode, asks the human to approve,\nexactly as for any task; for a documents-only change since the green suite it is\n`$STATE verify --scope step --files \"\u003Cdoc_tests>\" && $STATE verify --scope e2e && $STATE close`. Suggest `\u002Fclear` after it.\n\n## Switching\n\n- **To the standard loop (\"full\" from here):** `$STATE batch off --by \"\u003Cname>\" --why \"\u003Cwhy>\"`.\n  The standard budgets apply from now, counting what was spent — with two passes behind you,\n  the next verdict is already the human's. The session implements the remaining steps itself;\n  continue with `\u002Frr-build`, `\u002Frr-review`, `\u002Frr-ship`. Anyone may do this: it raises the rigor.\n- **To batch, mid-task:** lowers the rigor — the human's, in their own terminal:\n  `python3 \u003CAI_HOME>\u002Fskills\u002Fai-task\u002Fstate.py --root \u003Cproject> batch on --by \"\u003Cname>\"`. Never\n  from a tool call; the refusal (`APPROVAL_REFUSED`) prints that command.\n- **The project's default:** `batch.default` in `.ai\u002Fpolicies\u002Fweight.json`, beside\n  `batch.review_rounds_max`, `batch.max_suite_runs`, `batch.parallel` and `batch.tier`\n  (`\u002Frr-setup --update` adds the section). A policy is the human's and is frozen during a task.\n\n## Rules\n\n- The mode is the weight's and only rises; the scope guard, the two facts, the gate and the\n  three nevers are untouched by batch.\n- Three or more parallel agents run at BALANCED, never above — `start` refuses `batch.tier`\n  above it when `batch.parallel` is three or more; one reviewer per pass, at the mode's tier;\n  the session writes every conclusion (`model-routing.md`).\n- One writer of the state: a subagent never calls `state.py`; this session records every step,\n  run and verdict.\n- Failures and findings are fixed as batches, never one at a time; a test is never edited to\n  make it pass; what is outside the task is written down, not fixed.\n- No commit, push, merge or deploy. The task ends at `\u002Frr-ship`; approval is the human's.\n","# \u002Frr-build-all $ARGUMENTS\n\nOne task for the whole plan — the loop a developer otherwise runs by hand: implement every\nrow, test once, fix everything, review, fix everything, again, until the review passes. The\nmode (light, guarded, full) is still the weight's; batch only changes who implements, how\nfailures are fixed, and how many rounds run before the human. Resolve the install root once:\n\n```bash\nfor AI_HOME in \"${CLAUDE_CONFIG_DIR:-$HOME\u002F.claude}\" \"${CODEX_HOME:-$HOME\u002F.codex}\"; do\n  [ -d \"$AI_HOME\u002Fskills\u002Frr-build-all\" ] && break\ndone\nSTATE=\"python3 $AI_HOME\u002Fskills\u002Fai-task\u002Fstate.py\"\n```\n\nShared files resolve through `\"$AI_HOME\u002Friskrails\u002Fai-path\" \u003Cpath>`. Read `policies\u002Fsafety.md`\nand `using-riskrails.md` once per session. Every rule of `\u002Frr-build` holds here: nothing outside\nthe task's files, no commit, a refusal names the way out and is never reworded around.\n\n## 1. Resume or start\n\n```bash\n$STATE handoff --print\n```\n\n- **A task in flight whose handoff says `Batch: on`:** continue at its stage — §3 while\n  `steps --ready` prints a step, §4 after the last `step-done`, §5 after a green verify.\n- **A task in flight without that line:** it runs the standard way — say so and hand over to\n  `\u002Frr-build`. Only the human turns the loop on mid-task (`batch on`, §Switching).\n- **No task in flight:** §2.\n\n## 2. The plan → one task\n\n`$ARGUMENTS` is a programme plan (default: the newest `docs\u002Fsdlc\u002Fplans\u002F*.md`). Its spec's\n**Flagged concerns** are resolved or accepted, as `\u002Frr-plan` demands; a plan with an open\nblocking question is not started.\n\n1. **Rows → steps.** Every row of **Order of work** that an agent does becomes one step\n   `{\"id\", \"files\", \"tests\", \"after\"}`: the row's files, the tests it names, and in `after` the\n   rows it depends on — its *Consumes* line, a stated dependency, or an earlier row it builds\n   on. A refactoring row stays its own task (C3): leave it out and say so. A human row (G1, …)\n   is not a step: every row after it names in `after` every step before it, so nothing past\n   the human row is ready until the rows before it are done — and its question (3.) answered.\n2. **Weigh and start** — `\u002Frr-plan`'s \"A task\" section, on the union of the steps' files, then:\n\n   ```bash\n   $STATE start --goal \"\u003Cthe plan's title>\" --workflow feature --files \"\u003Cevery file of every step>\" \\\n     --tests \"\u003Cevery test>\" [--declare …] --plan-file \u002Fpath\u002Fsteps.json \\\n     --batch --by \"\u003Cthe human's name>\" --why \"\u002Frr-build-all docs\u002Fsdlc\u002Fplans\u002F\u003Cslug>.md\" \\\n     [--mode \u003Chigher> --by \"\u003Cname>\"]\n   ```\n\n   An `after` id must exist and form no cycle; two steps that name one file are ordered by\n   `start` itself (`ordered: s4 after s1 (shares …)`), so two implementers never hold one file.\n   In full mode the written plan is the programme plan the human already accepted. With\n   `batch.default` true in `.ai\u002Fpolicies\u002Fweight.json` the flag is implied; `--no-batch` opts out.\n3. **A human row** becomes a question the moment every step before it is done — asked\n   **before** the next wave is dispatched:\n   `$STATE ask \"G1 is yours: \u003Cwhat>\" --option \"A: done\" --option \"B: not yet\" --recommend A`,\n   then **stop**. Never dispatch while a question is pending (`step-done` and `close` exit 4;\n   `steps --ready` still answers); `\u002Frr-build-all --resume` goes on once the human answered.\n\n## 3. Implement, by waves\n\n```bash\n$STATE steps --ready               # the next wave: the steps whose dependencies are done, one id per line\n$STATE steps --json                # batch_policy.tier and .parallel: the implementers' tier, how many at once\n$STATE profile --tier \u003Cthat tier>  # the tier's model in this runtime (BALANCED as shipped)\n```\n\n**Resuming mid-wave** (after a compaction or a stop): before dispatching, compare\n`git status --short` with each ready step's files. A ready step whose files already changed\nwas dispatched before and never recorded — do not send it again blind: brief its implementer\nwith that partial diff to finish, or `step-done` it if its tests pass.\n\nDispatch one subagent per ready step, at most `batch.parallel` (shipped 3) at once — under\nClaude Code a `general-purpose` agent with `model:` set to that tier's model; under Codex a\nspawn with no agent named, on that model. Its brief is the implementer contract,\n`agents\u002Fimplementer.md` in the shared tree, filled in:\n\n- **Goal** — the row's text from the plan. **Allowed files** — the step's files, nothing else.\n  **Forbidden files** — every other step's files, \"another implementer holds them\".\n  **Required behaviour \u002F behaviour that must not change** — from the row and the spec.\n  **Required tests** — the step's tests.\n- Its own tests, run directly: `step_test_command` from `.ai\u002Fpolicies\u002Ftesting.md` with the\n  step's tests in place of `{files}`. **A subagent never calls `state.py`** — one state file,\n  one writer, this session.\n- It returns `## RESULT`, or `## SCOPE_CHANGE_REQUIRED`, or `## QUESTIONS_NEEDED`.\n\nAfter the wave, in this session:\n\n- `RESULT` → `$STATE step-done \u003Cid>` for each step. Exit 6 names a file outside the task:\n  `$STATE start --amend --files \"+\u003Cpath>\" --why \"…\"`, then `step-done` again.\n- `SCOPE_CHANGE_REQUIRED` → amend as above and send the same step again. A file that belongs\n  to another change goes into `.ai\u002Fproject\u002Fknown-risks.md`, not into the task.\n- `QUESTIONS_NEEDED` → `$STATE ask --batch \u002Fpath\u002Fquestions.json`, show the questions, **stop**;\n  the human answers, and `\u002Frr-build-all --resume` goes on from the handoff.\n\nThen `steps --ready` again. Nothing ready and nothing pending: §4. Keep this session small:\nread `## RESULT` blocks, never a subagent's diff; `git status --short` is the only view of the\ntree you need between waves.\n\n## 4. Tests once, failures fixed as one parallel batch\n\nAfter the last wave, `$STATE step-done \u003Cid> && $STATE verify --scope all`:\n\n```bash\n$STATE verify --scope all\n$STATE verify --scope all --e2e not-applicable \"\u003Cwhy the change reaches no e2e flow>\"\n```\n\nGreen: §5. **Red:** a FAST `log-reader` reads the log by path and returns every failure\nclassified — new regression, existing failure, environment, unknown (`testing.md`, \"Reading a\nfailure\") — each with `file:line`. Then:\n\n- an existing failure → one entry in `.ai\u002Fproject\u002Fknown-risks.md`; an environment failure →\n  one more run, no code change;\n- the new regressions, grouped by the file they point at → one fixer subagent per group, in\n  parallel, at most `batch.parallel`, on the batch tier, briefed with the failures' excerpts\n  and the files it may touch (that group's files, inside the task's). Fix the code, never the\n  test; report `files_changed` and what stayed unfixed;\n- one run: `$STATE verify --scope all --after-fix \"\u003Cn> regressions: \u003Cone line each>\"`.\n\nRed again → the same once more. Exit 6 `SUITE_BUDGET` — past `batch.max_suite_runs`\n(shipped 8) — is the human's run: the pick in the dialog, or the command it prints, in their\nown terminal. **Stop.**\n\n## 5. Review, findings fixed as one parallel batch, again until it passes\n\nOne `ai-reviewer` at the mode's tier, briefed as `\u002Frr-review` §2 says: the whole diff since the\ntask's base, the task's files, the tests, the last verify log; unbounded \"look for\" in full\nmode, the security focus when the domain axis is 2. Record it verbatim:\n\n```bash\n$STATE review --verdict \u003Cpass|pass_with_findings|blockers_open> --tier \u003CBALANCED|STRONG> \\\n  [--focus security] --report - \u003C\u003C'EOF'\n\u003Cthe reviewer's FINDINGS and VERDICT, unchanged>\nEOF\n```\n\n- **pass** → §6.\n- **pass_with_findings** → the in-scope HIGHs, grouped by file, one fixer per group in\n  parallel; `$STATE verify --scope all --after-fix \"H1,H2: …\"`; `$STATE review --closes \"H1,H2\"`\n  → §6. MEDIUM and LOW: `memory risk add` as `\u002Frr-review` §4 says, for every verdict; only a\n  BLOCKER, a HIGH or a regression buys an `--after-fix` run.\n- **blockers_open** → every in-scope BLOCKER and HIGH, grouped by file, one fixer per group in\n  parallel; `$STATE verify --scope all --after-fix \"B1,H1: …\"` (red: §4's batch); then\n  **§5 again, a full pass** — a fix batch across many files is a new change, not a scoped one.\n\nExit 6 `REVIEW_BUDGET` after `batch.review_rounds_max` passes (shipped 3 — three failed\nreviews) — the next verdict is the human's: the pick in the dialog, or the command it prints,\nin their own terminal. Show the open findings and **stop**. A `verify` past its budget inside\nthis loop is `SUITE_BUDGET`, handled the same way.\n\n## 6. Ship\n\n`\u002Frr-ship` — `close` measures the two facts and, in full mode, asks the human to approve,\nexactly as for any task; for a documents-only change since the green suite it is\n`$STATE verify --scope step --files \"\u003Cdoc_tests>\" && $STATE verify --scope e2e && $STATE close`. Suggest `\u002Fclear` after it.\n\n## Switching\n\n- **To the standard loop (\"full\" from here):** `$STATE batch off --by \"\u003Cname>\" --why \"\u003Cwhy>\"`.\n  The standard budgets apply from now, counting what was spent — with two passes behind you,\n  the next verdict is already the human's. The session implements the remaining steps itself;\n  continue with `\u002Frr-build`, `\u002Frr-review`, `\u002Frr-ship`. Anyone may do this: it raises the rigor.\n- **To batch, mid-task:** lowers the rigor — the human's, in their own terminal:\n  `python3 \u003CAI_HOME>\u002Fskills\u002Fai-task\u002Fstate.py --root \u003Cproject> batch on --by \"\u003Cname>\"`. Never\n  from a tool call; the refusal (`APPROVAL_REFUSED`) prints that command.\n- **The project's default:** `batch.default` in `.ai\u002Fpolicies\u002Fweight.json`, beside\n  `batch.review_rounds_max`, `batch.max_suite_runs`, `batch.parallel` and `batch.tier`\n  (`\u002Frr-setup --update` adds the section). A policy is the human's and is frozen during a task.\n\n## Rules\n\n- The mode is the weight's and only rises; the scope guard, the two facts, the gate and the\n  three nevers are untouched by batch.\n- Three or more parallel agents run at BALANCED, never above — `start` refuses `batch.tier`\n  above it when `batch.parallel` is three or more; one reviewer per pass, at the mode's tier;\n  the session writes every conclusion (`model-routing.md`).\n- One writer of the state: a subagent never calls `state.py`; this session records every step,\n  run and verdict.\n- Failures and findings are fixed as batches, never one at a time; a test is never edited to\n  make it pass; what is outside the task is written down, not fixed.\n- No commit, push, merge or deploy. The task ends at `\u002Frr-ship`; approval is the human's.\n",{},"main","84a91bd082ae3a289fcbe8a0f1a52bdebe5f638a","skills\u002Frr-build-all",0,"https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module","https:\u002F\u002Fvanssata.github.io\u002Fclaude-agentic-sdlc-module\u002F","https:\u002F\u002Fgithub.com\u002Fvanssata","2026-11-14 21:12:58",{"source":7,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add vanssata\u002Fclaude-agentic-sdlc-module --skill rr-build-all",{"tool":37,"command":38},"bunx","bunx skills add vanssata\u002Fclaude-agentic-sdlc-module --skill rr-build-all",{"tool":40,"command":41},"pnpm","pnpm dlx skills add vanssata\u002Fclaude-agentic-sdlc-module --skill rr-build-all","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Fskills\u002Frr-build-all\u002FSKILL.md",true,[],[46,53,60,66,73,80],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":18},"01M7CF1FY7PYAK5378YCV5RV5S","sylius-fixtures","Load the Sylius fixture suite into the local database","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsetup\u002Fnew-shop-like\u002Fdot-claude\u002Fskills\u002Fsylius-fixtures\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":18},"01M7CF1CTV8SNSJEGJ88HFM6JC","deploy-staging","Deploy the current branch to staging and smoke-test the checkout. Use on \"deploy to staging\".","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsetup\u002Fnew-shop-like\u002Fdot-claude\u002Fskills\u002Fdeploy-staging\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":18},"01M7CF1ARVD7MZGDM7FJKE8CNR","rr-status","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Fskills\u002Frr-status\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M7CF136MTETD8RF1NYJGYHK7","rr-spec","Turn a rough idea into one file, docs\u002Fsdlc\u002Fspecs\u002F\u003Cslug>.md — an `## Intent` section (problem, outcome, affected users and systems, constraints, decisions taken) from a short one-question-at-a-time brainstorm recorded through state.py, then requirements and design from the architect agent on the STRONG tier, checked against the project's constitution, instruction files, skills and ADRs, ending with explicitly flagged concerns (especially contradicting policies). Accepts an existing docs\u002Fsdlc\u002Fintent\u002F\u003Cslug>.md instead of the brainstorm. Use when starting a feature, \"write the spec\", \"let's define what we want\", \"design this\", before any plan or code.","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Fskills\u002Frr-spec\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":18},"01M7CF0WJVAB1ZCC0GF6NZ7H87","rr-ship","Close the task — state.py close measures the two facts (a green whole-suite verify on the tree as it is now; every changed file inside the task's files), applies the mode table and either archives the task or, in full mode, requests the human's approval and prints the exact approve command for their own terminal. Writes the commit message body and suggests the commit; never merges or deploys, and commits, pushes or approves only as the command the human picked in the dialog, verbatim, once. Use on \"\u002Frr-ship\", \"close the task\", \"ship it\", \"ready for approval\", after \u002Frr-build or \u002Frr-review.","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Fskills\u002Frr-ship\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M7CF0SZAWZFAJBEK64M27KRF","rr-setup","https:\u002F\u002Fgithub.com\u002Fvanssata\u002Fclaude-agentic-sdlc-module\u002Fblob\u002Fmain\u002Fskills\u002Frr-setup\u002FSKILL.md",[],[]]