[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M758MD74ZF37RAATF4B5D0N2":3,"skill-same:01M758MD74ZF37RAATF4B5D0N2":62},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":13,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":6,"updatedAt":30,"install":31,"resources":43},"01M758MD74ZF37RAATF4B5D0N2","reflect",null,"Spawn three parallel review subagents over the active transcript, surface learnings, and route each to a concrete edit on an existing skill. Use when the user says reflect.","BainaryTD\u002Fbainary-skill","BainaryTD","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F121595278?v=4","01M758GV0FS7DF77TMEYD2JPWJ","01M758GV1EY6NXCX50681D0VV4",0,"MIT","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Freflect\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: reflect\ndescription: Spawn three parallel review subagents over the active transcript, surface learnings, and route each to a concrete edit on an existing skill. Use when the user says reflect.\ndisable-model-invocation: true\n---\n\n# Reflect\n\nMine the current conversation for durable learnings, then route them into skill edits.\n\n## When to invoke\n\nInvoke when the user says \"reflect\" or \"\u002Freflect\". Skip when the conversation is trivial, off-topic, or already covered by an existing skill the parent followed correctly. One-offs are not learnings.\n\n## Process\n\n### 1. Locate the active transcript\n\nThe parent finds its own transcript file before fanning out. Use only the active workspace's transcript directory. Never glob across every project's directory. That crosses workspace boundaries and reads private chats from unrelated projects.\n\nWhere the transcript lives depends on your harness:\n\n- **Cursor:** the system prompt names the active workspace's `agent-transcripts\u002F` directory. Use that path.\n\n  ```bash\n  ls -t \u003Cagent-transcripts>\u002F*.jsonl \u003Cagent-transcripts>\u002F*\u002F*.jsonl \u003Cagent-transcripts>\u002F*\u002Fsubagents\u002F*.jsonl 2>\u002Fdev\u002Fnull | head -10\n  ```\n\n  Three transcript layouts: legacy flat (`\u003Cid>.jsonl`), current nested (`\u003Cid>\u002F\u003Cid>.jsonl`), and subagent (`\u003Cparent>\u002Fsubagents\u002F\u003Cchild>.jsonl`).\n- **Claude Code:** `~\u002F.claude\u002Fprojects\u002F\u003Cslug>\u002F\u003Csession-id>.jsonl`, where `\u003Cslug>` is the workspace path with every character that isn't a letter or digit turned into \"-\".\n- **Pi:** `~\u002F.pi\u002Fagent\u002Fsessions\u002F--\u003Cslug>--\u002F*.jsonl`, where `\u003Cslug>` is the workspace path with the leading slash dropped and each \"\u002F\" turned into \"-\".\n- **Codex:** `~\u002F.codex\u002Fsessions\u002FYYYY\u002FMM\u002FDD\u002Frollout-*.jsonl`. Keep only files whose first line has `payload.cwd` equal to the workspace path.\n- **Other harnesses:** check the harness's session directory for this workspace.\n\nTake the newest candidates first. Confirm a candidate by finding the conversation's opening user prompt in its first user message. Take the matching path. If no path resolves, write a tight digest of the session and pass that instead.\n\n### 2. Spawn three reviewers in parallel\n\nOne message, three `Task` calls, `subagent_type: generalPurpose`, with `model` set as below, agent mode (`readonly: false`). Reviewers need MCP access for context lookups (tickets, chat threads, observability traces referenced in the transcript). Readonly strips MCPs.\n\n**Other harnesses.** The spawns in this skill use Cursor's `Task` tool. In another harness, use its subagent tool: `Agent` in Claude Code (`subagent_type: general-purpose`), `task` in OpenCode (`subagent_type: general`), `spawn_agent` in Codex. Keep the prompt and the model. Drop parameters your tool doesn't have. If your harness has no subagent tool, as in Pi without an extension, run each role yourself, one after another.\n\nEach reviewer and the synthesizer name a role line in the pstack settings file (`~\u002F.cursor\u002Frules\u002Fpstack-models.mdc` in Cursor, `~\u002F.agents\u002Fpstack-models.md` in other harnesses) and a default. Set `model` to that line's value, or to the default if the file or the line is missing. Leave `model` unset when the value is `auto` or `inherit-parent`. If your subagent tool rejects a slug, use the default and say so. If it rejects the default, use the closest valid slug of the same family from its error message or your harness's model list.\n\n| Lens | Role line | Default `model` | Prompt template |\n|---|---|---|---|\n| Judgment | `reflect judgment, divergent, synthesizer` | `claude-opus-5-5-max` | `references\u002Fjudgment-reviewer.md` |\n| Tooling | `reflect tooling` | `gpt-5.6-sol-max` | `references\u002Ftooling-reviewer.md` |\n| Divergent | `reflect judgment, divergent, synthesizer` | `claude-opus-5-5-max` | `references\u002Fdivergent-reviewer.md` |\n\nPass each template verbatim, substituting the transcript path or digest where marked. Reviewers return findings in the `Task` response body.\n\n### 3. Synthesize\n\nOne `Task` call, `subagent_type: generalPurpose`, with `model` from the `reflect judgment, divergent, synthesizer` line (default `claude-opus-5-5-max`), agent mode (`readonly: false`). The synthesizer's quality check includes spot-verifying citations, which can require MCP access. Readonly strips MCPs. Use `references\u002Fsynthesizer.md` verbatim, with each reviewer's full output inlined where marked. The synthesizer returns a structured Accepted \u002F Rejected \u002F Backlog list.\n\n### 4. Structural enforcement check\n\nSanity-check the synthesizer's Accepted list. For any item that would be enforced more reliably by a lint rule, script, metadata flag, or runtime check, move it from Accepted to Backlog. See the **encode-lessons-in-structure** principle skill.\n\n### 5. Apply\n\nBefore applying any Accepted edit, present the synthesizer's full Accepted\u002FRejected\u002FBacklog output to the user and wait for explicit approval. The user picks which subset to apply and may redirect routings. Skill changes affect every future agent in the org. Do not auto-apply.\n\nBacklog items file to whatever devex \u002F backlog tracker your team uses automatically. Only the Accepted list waits for approval.\n\nFor each approved Accepted item, follow the Routing field exactly:\n\n- Trivial existing-skill edit (a one-line bullet, a tightened sentence, a stale fact corrected): parent does directly.\n- Substantive existing-skill edit (a new section, a new pattern table, more than ~10 lines): hand to your harness's skill-authoring skill and run its draft \u002F test \u002F iterate loop. That is `create-skill` in Cursor (built in) or Anthropic's `skill-creator`. If you have neither, follow the Agent Skills format at agentskills.io. \"`create-skill`\" below means whichever of these you use.\n- `tune description: \u003Cskill path>` (the skill exists but didn't trigger when it should have): hand to `create-skill` and run its description-optimization loop.\n- `new skill via create-skill: \u003Ckebab-name>`: hand creation to `create-skill`. Do not invent the shape ad hoc.\n\nIf your environment ships a SKILL.md validator, run it on every touched skill before declaring done. Skip this step if it doesn't.\n\n### 6. Summarize for the user\n\nShort list, no preamble:\n\n- Edits applied: `\u003Cskill path>`. What changed, one line each.\n- New skills created: `\u003Cskill path>`. One line each (rare).\n- Backlog filed to the devex tracker: `\u003Cissue title>` (`\u003Ctags>`). One line each.\n- Dropped: one line per rejected finding + reason from the synthesizer.\n","# Reflect\n\nMine the current conversation for durable learnings, then route them into skill edits.\n\n## When to invoke\n\nInvoke when the user says \"reflect\" or \"\u002Freflect\". Skip when the conversation is trivial, off-topic, or already covered by an existing skill the parent followed correctly. One-offs are not learnings.\n\n## Process\n\n### 1. Locate the active transcript\n\nThe parent finds its own transcript file before fanning out. Use only the active workspace's transcript directory. Never glob across every project's directory. That crosses workspace boundaries and reads private chats from unrelated projects.\n\nWhere the transcript lives depends on your harness:\n\n- **Cursor:** the system prompt names the active workspace's `agent-transcripts\u002F` directory. Use that path.\n\n  ```bash\n  ls -t \u003Cagent-transcripts>\u002F*.jsonl \u003Cagent-transcripts>\u002F*\u002F*.jsonl \u003Cagent-transcripts>\u002F*\u002Fsubagents\u002F*.jsonl 2>\u002Fdev\u002Fnull | head -10\n  ```\n\n  Three transcript layouts: legacy flat (`\u003Cid>.jsonl`), current nested (`\u003Cid>\u002F\u003Cid>.jsonl`), and subagent (`\u003Cparent>\u002Fsubagents\u002F\u003Cchild>.jsonl`).\n- **Claude Code:** `~\u002F.claude\u002Fprojects\u002F\u003Cslug>\u002F\u003Csession-id>.jsonl`, where `\u003Cslug>` is the workspace path with every character that isn't a letter or digit turned into \"-\".\n- **Pi:** `~\u002F.pi\u002Fagent\u002Fsessions\u002F--\u003Cslug>--\u002F*.jsonl`, where `\u003Cslug>` is the workspace path with the leading slash dropped and each \"\u002F\" turned into \"-\".\n- **Codex:** `~\u002F.codex\u002Fsessions\u002FYYYY\u002FMM\u002FDD\u002Frollout-*.jsonl`. Keep only files whose first line has `payload.cwd` equal to the workspace path.\n- **Other harnesses:** check the harness's session directory for this workspace.\n\nTake the newest candidates first. Confirm a candidate by finding the conversation's opening user prompt in its first user message. Take the matching path. If no path resolves, write a tight digest of the session and pass that instead.\n\n### 2. Spawn three reviewers in parallel\n\nOne message, three `Task` calls, `subagent_type: generalPurpose`, with `model` set as below, agent mode (`readonly: false`). Reviewers need MCP access for context lookups (tickets, chat threads, observability traces referenced in the transcript). Readonly strips MCPs.\n\n**Other harnesses.** The spawns in this skill use Cursor's `Task` tool. In another harness, use its subagent tool: `Agent` in Claude Code (`subagent_type: general-purpose`), `task` in OpenCode (`subagent_type: general`), `spawn_agent` in Codex. Keep the prompt and the model. Drop parameters your tool doesn't have. If your harness has no subagent tool, as in Pi without an extension, run each role yourself, one after another.\n\nEach reviewer and the synthesizer name a role line in the pstack settings file (`~\u002F.cursor\u002Frules\u002Fpstack-models.mdc` in Cursor, `~\u002F.agents\u002Fpstack-models.md` in other harnesses) and a default. Set `model` to that line's value, or to the default if the file or the line is missing. Leave `model` unset when the value is `auto` or `inherit-parent`. If your subagent tool rejects a slug, use the default and say so. If it rejects the default, use the closest valid slug of the same family from its error message or your harness's model list.\n\n| Lens | Role line | Default `model` | Prompt template |\n|---|---|---|---|\n| Judgment | `reflect judgment, divergent, synthesizer` | `claude-opus-5-5-max` | `references\u002Fjudgment-reviewer.md` |\n| Tooling | `reflect tooling` | `gpt-5.6-sol-max` | `references\u002Ftooling-reviewer.md` |\n| Divergent | `reflect judgment, divergent, synthesizer` | `claude-opus-5-5-max` | `references\u002Fdivergent-reviewer.md` |\n\nPass each template verbatim, substituting the transcript path or digest where marked. Reviewers return findings in the `Task` response body.\n\n### 3. Synthesize\n\nOne `Task` call, `subagent_type: generalPurpose`, with `model` from the `reflect judgment, divergent, synthesizer` line (default `claude-opus-5-5-max`), agent mode (`readonly: false`). The synthesizer's quality check includes spot-verifying citations, which can require MCP access. Readonly strips MCPs. Use `references\u002Fsynthesizer.md` verbatim, with each reviewer's full output inlined where marked. The synthesizer returns a structured Accepted \u002F Rejected \u002F Backlog list.\n\n### 4. Structural enforcement check\n\nSanity-check the synthesizer's Accepted list. For any item that would be enforced more reliably by a lint rule, script, metadata flag, or runtime check, move it from Accepted to Backlog. See the **encode-lessons-in-structure** principle skill.\n\n### 5. Apply\n\nBefore applying any Accepted edit, present the synthesizer's full Accepted\u002FRejected\u002FBacklog output to the user and wait for explicit approval. The user picks which subset to apply and may redirect routings. Skill changes affect every future agent in the org. Do not auto-apply.\n\nBacklog items file to whatever devex \u002F backlog tracker your team uses automatically. Only the Accepted list waits for approval.\n\nFor each approved Accepted item, follow the Routing field exactly:\n\n- Trivial existing-skill edit (a one-line bullet, a tightened sentence, a stale fact corrected): parent does directly.\n- Substantive existing-skill edit (a new section, a new pattern table, more than ~10 lines): hand to your harness's skill-authoring skill and run its draft \u002F test \u002F iterate loop. That is `create-skill` in Cursor (built in) or Anthropic's `skill-creator`. If you have neither, follow the Agent Skills format at agentskills.io. \"`create-skill`\" below means whichever of these you use.\n- `tune description: \u003Cskill path>` (the skill exists but didn't trigger when it should have): hand to `create-skill` and run its description-optimization loop.\n- `new skill via create-skill: \u003Ckebab-name>`: hand creation to `create-skill`. Do not invent the shape ad hoc.\n\nIf your environment ships a SKILL.md validator, run it on every touched skill before declaring done. Skip this step if it doesn't.\n\n### 6. Summarize for the user\n\nShort list, no preamble:\n\n- Edits applied: `\u003Cskill path>`. What changed, one line each.\n- New skills created: `\u003Cskill path>`. One line each (rare).\n- Backlog filed to the devex tracker: `\u003Cissue title>` (`\u003Ctags>`). One line each.\n- Dropped: one line per rejected finding + reason from the synthesizer.\n",{"name":5,"description":7,"disable-model-invocation":23},true,"main","6e59af6cd45e52f0f16c811134303a409e90292a","skills\u002Fpstack\u002Freflect","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill","https:\u002F\u002Fgithub.com\u002FBainaryTD","https:\u002F\u002Fcodecrafthorizon.com","2026-10-06 12:07:04",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":23},[33,36,39],{"tool":34,"command":35},"npx","npx skills add BainaryTD\u002Fbainary-skill --skill reflect",{"tool":37,"command":38},"bunx","bunx skills add BainaryTD\u002Fbainary-skill --skill reflect",{"tool":40,"command":41},"pnpm","pnpm dlx skills add BainaryTD\u002Fbainary-skill --skill reflect","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Freflect\u002FSKILL.md",[44,50,54,58],{"path":45,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":48,"sourceHtmlUrl":49},"references\u002Fdivergent-reviewer.md","REFERENCE","参考文档",4186,"https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Freflect\u002Freferences\u002Fdivergent-reviewer.md",{"path":51,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":52,"sourceHtmlUrl":53},"references\u002Fjudgment-reviewer.md",3552,"https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Freflect\u002Freferences\u002Fjudgment-reviewer.md",{"path":55,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":56,"sourceHtmlUrl":57},"references\u002Fsynthesizer.md",4164,"https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Freflect\u002Freferences\u002Fsynthesizer.md",{"path":59,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":60,"sourceHtmlUrl":61},"references\u002Ftooling-reviewer.md",4906,"https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Freflect\u002Freferences\u002Ftooling-reviewer.md",[63,70,77,84,91,98],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":18},"01M758N5F1RT27WBHE9JR4693P","why","Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat, infrastructure observability, error tracking, product analytics warehouse) in parallel, then returns a cited read on decisions and tradeoffs. Use how for runtime behavior.","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Fwhy\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":18},"01M758N3PPB7JET89EQ3AFF188","unslop","Cut AI tells from any writing. Must always apply.","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Funslop\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":18},"01M758N26N2FV8AHEEXK8XY55E","typescript-best-practices","TypeScript best practices. Use when reading or editing any .ts or .tsx file.","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Ftypescript-best-practices\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":18},"01M758MZC2N6ZWDV2APPZ2RVM0","technical-writing","Layered technical-writing standard: Diátaxis structure, Google developer style sentences, STE instruction rules, Global English syntax. Use for \u002Ftechnical-writing or when writing or reviewing docs, RFCs, readmes, PR descriptions, or commit messages.","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Ftechnical-writing\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":18},"01M758MVGBVV3HSMS1S6RHFXQN","teach","Explain a body of work plainly so a person actually understands it. Runs the `how` and `why` skills and weaves what they find into one clear explanation. Use for 'teach me this', 'help me really understand X', 'explain this change or subsystem to me'.","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Fteach\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":18},"01M758MQ808137A88YPFJRCMHR","tdd","Use only when the user explicitly asks for TDD, a failing test, or a regression test, OR when the bug has an obvious cheap local test target. Skip when the test path is unclear, expensive, integration-heavy, or not requested.","https:\u002F\u002Fgithub.com\u002FBainaryTD\u002Fbainary-skill\u002Fblob\u002Fmain\u002Fskills\u002Fpstack\u002Ftdd\u002FSKILL.md",[],[]]